Loading ...
Sorry, an error occurred while loading the content.

RE: [Behavioral-Finance] Dow Monthly returns

Expand Messages
  • Esha Malik
      ... From: Bob Bronson Subject: RE: [Behavioral-Finance] Dow Monthly returns To: Behavioral-Finance@yahoogroups.com Date: Monday,
    Message 1 of 15 , Sep 6, 2010
    View Source
    • 0 Attachment
       


      --- On Mon, 9/6/10, Bob Bronson <bob@...> wrote:

      From: Bob Bronson <bob@...>
      Subject: RE: [Behavioral-Finance] Dow Monthly returns
      To: Behavioral-Finance@yahoogroups.com
      Date: Monday, September 6, 2010, 9:04 AM

       

      Based upon the Dow’s price-only history as Mark has referenced below, when the year-to-date

      through August was down, as was the situation this year, September has been down about twice

      as often as it has been up.

       

      However, using total return (dividends reinvested) with our monthly CMI database for the 139

      years since 1870 (S&P 90 and 500 indexes starting in Jan ’26 and Jan ’50, respectively, and

      S&P-Wilshire 5000 since Apr ’01) – available upon request – there have been 45 instances with

      their years-to-date down through August, of which 28, or 62%, Septembers had negative returns.

       

      Furthermore, during the 12 second Presidential-term years since 1962, when the 48-month,

      socio-political-economic, or Kitchin, cycle fully emerged (see footnotes # 4, 10, 14 and 20

      here:  A Forecasting Model That Integrates Multiple Business and Stock Market Cycles

      September has been down 67% of the time, or twice as often as it has been up: eight times

      down compared to only four times up. 

       

      Considering only Septembers when both it was a Presidential term second year, typically the

      weakest of the four years (the political discord and economic uncertainty this year is unusually

      bearish) and the stock market, measured by our CMI, was down year-to-date through August,

      five of those seven instances were down or 71% of the time.

       

      Finally, four of those seven were during Supercycle Bear Market Periods (e.g., 1881-1896,

      1906-21, 1929-49, 1966-82 and 2000 to date) of which three, or 75%, of those Septembers

      had negative returns - 1966, 1974 and 2002 , while September 2006 gained 2.3%.  

       

      Since our work clearly shows that we’re still in a Supercycle Bear Market Period, in particular,

      the most devastating of the two kinds: a deflationary economic Supercycle Winter – see the

      summary of our quantified Supercycle Economic Seasons below - we believe the odds that

      this September, while it has started off with a strong start, will decline for the rest of the month,

      and it will probably even finish with a net negative return. 

       

      Of course, we have other reasons for expecting this also – see the summary of our forecasting

      models further below.

       

      Bob Bronson

      Bronson Capital Markets Research

      http://financialsense.com/editorials/bronson/main.html

      http://www.financialsense.com/contributors/bob-bronson

       

       

       

       

      http://www.financialsense.com/sites/default/files/users/u188/images/2010/supercycle-economic-seasons.jpg

       

      http://www.financialsense.com/sites/default/files/users/u188/images/2010/bronson-forecast-models.jpg

       

       

      -----Original Message-----
      From: Behavioral-Finance@yahoogroups.com [mailto:Behavioral-Finance@yahoogroups.com] On Behalf Of mark
      Sent: Sunday, September 05, 2010 1:49 PM
      To: Behavioral-Finance@yahoogroups.com
      Subject: [Behavioral-Finance] Dow Monthly returns

       

      This chart shows the Dow Industrials returns on a monthly basis back to 1886.  It is sorted on September returns

      http://www.mutualfundsystem.com/newsletters/DowMonthlyReturns.pdf

       

       

       

      ------------------------------------

       

      you may unsubscribe by sending an email to

       

      Behavioral-Finance-unsubscribe@yahoogroups.com

      Yahoo! Groups Links

       

      <*> To visit your group on the web, go to:

          http://groups.yahoo.com/group/Behavioral-Finance/

       

      <*> Your email settings:

          Individual Email | Traditional

       

      <*> To change settings online go to:

          http://groups.yahoo.com/group/Behavioral-Finance/join

          (Yahoo! ID required)

       

      <*> To change settings via email:

          Behavioral-Finance-digest@yahoogroups.com

          Behavioral-Finance-fullfeatured@yahoogroups.com

       

      <*> To unsubscribe from this group, send an email to:

          Behavioral-Finance-unsubscribe@yahoogroups.com

       

      <*> Your use of Yahoo! Groups is subject to:

          http://docs.yahoo.com/info/terms/


    • SergeyTS
      Hello, This is Annual cycle for Dow: the price history since 1789 year is used. Actually, Dow exists since 1885, before that the data are the interpolated
      Message 2 of 15 , Sep 6, 2010
      View Source
      • 0 Attachment
        Hello,
         
        This is Annual cycle for Dow:
         
         
         
        the price history since 1789 year is used. Actually, Dow exists since 1885, before that the data are the interpolated prices of the most  liquid companies.
         
        Best regards.
        Sergey.
         
         
        ----- Original Message -----
        Sent: Monday, September 06, 2010 5:04 AM
        Subject: RE: [Behavioral-Finance] Dow Monthly returns

         

        Based upon the Dow’s price-only history as Mark has referenced below, when the year-to-date

        through August was down, as was the situation this year, September has been down about twice

        as often as it has been up.

        However, using total return (dividends reinvested) with our monthly CMI database for the 139

        years since 1870 (S&P 90 and 500 indexes starting in Jan ’26 and Jan ’50, respectively, and

        S&P-Wilshire 5000 since Apr ’01) – available upon request – there have been 45 instances with

        their years-to-date down through August, of which 28, or 62%, Septembers had negative returns.

        Furthermore, during the 12 second Presidential-term years since 1962, when the 48-month,

        socio-political-economic, or Kitchin, cycle fully emerged (see footnotes # 4, 10, 14 and 20

        here:  A Forecasting Model That Integrates Multiple Business and Stock Market Cycles

        September has been down 67% of the time, or twice as often as it has been up: eight times

        down compared to only four times up. 

        Considering only Septembers when both it was a Presidential term second year, typically the

        weakest of the four years (the political discord and economic uncertainty this year is unusually

        bearish) and the stock market, measured by our CMI, was down year-to-date through August,

        five of those seven instances were down or 71% of the time.

        Finally, four of those seven were during Supercycle Bear Market Periods (e.g., 1881-1896,

        1906-21, 1929-49, 1966-82 and 2000 to date) of which three, or 75%, of those Septembers

        had negative returns - 1966, 1974 and 2002 , while September 2006 gained 2.3%.  

        Since our work clearly shows that we’re still in a Supercycle Bear Market Period, in particular,

        the most devastating of the two kinds: a deflationary economic Supercycle Winter – see the

        summary of our quantified Supercycle Economic Seasons below - we believe the odds that

        this September, while it has started off with a strong start, will decline for the rest of the month,

        and it will probably even finish with a net negative return. 

        Of course, we have other reasons for expecting this also – see the summary of our forecasting

        models further below.

        Bob Bronson

        Bronson Capital Markets Research

        http://financialsense.com/editorials/bronson/main.html

        http://www.financialsense.com/contributors/bob-bronson

        http://www.financialsense.com/sites/default/files/users/u188/images/2010/supercycle-economic-seasons.jpg

        http://www.financialsense.com/sites/default/files/users/u188/images/2010/bronson-forecast-models.jpg

        -----Original Message-----
        From: Behavioral-Finance@yahoogroups.com [mailto:Behavioral-Finance@yahoogroups.com] On Behalf Of mark
        Sent: Sunday, September 05, 2010 1:49 PM
        To: Behavioral-Finance@yahoogroups.com
        Subject: [Behavioral-Finance] Dow Monthly returns

        This chart shows the Dow Industrials returns on a monthly basis back to 1886.  It is sorted on September returns

        http://www.mutualfundsystem.com/newsletters/DowMonthlyReturns.pdf

        ------------------------------------

        you may unsubscribe by sending an email to

        Behavioral-Finance-unsubscribe@yahoogroups.com

        Yahoo! Groups Links

        <*> To visit your group on the web, go to:

            http://groups.yahoo.com/group/Behavioral-Finance/

        <*> Your email settings:

            Individual Email | Traditional

        <*> To change settings online go to:

            http://groups.yahoo.com/group/Behavioral-Finance/join

            (Yahoo! ID required)

        <*> To change settings via email:

            Behavioral-Finance-digest@yahoogroups.com

            Behavioral-Finance-fullfeatured@yahoogroups.com

        <*> To unsubscribe from this group, send an email to:

            Behavioral-Finance-unsubscribe@yahoogroups.com

        <*> Your use of Yahoo! Groups is subject to:

            http://docs.yahoo.com/info/terms/

      • MarkK
        Since there are numerous cycles involved what one are you using? Price? Economic? President? Elliot Wave? Hurts? And the list continues MarkK From:
        Message 3 of 15 , Sep 6, 2010
        View Source
        • 0 Attachment

          Since there are numerous cycles involved what one are you using?
          Price? Economic? President? Elliot Wave? Hurts? And the list continues

           

           

          MarkK

           

           

           

           

          From: Behavioral-Finance@yahoogroups.com [mailto:Behavioral-Finance@yahoogroups.com] On Behalf Of SergeyTS
          Sent: Monday, September 06, 2010 9:49 AM
          To: Behavioral-Finance@yahoogroups.com
          Subject: Re: [Behavioral-Finance] Dow Monthly returns

           

           

          Hello,

           

          This is Annual cycle for Dow:

           

           

           

          the price history since 1789 year is used. Actually, Dow exists since 1885, before that the data are the interpolated prices of the most  liquid companies.

           

          Best regards.

          Sergey.

           

           

          ----- Original Message -----

          Sent: Monday, September 06, 2010 5:04 AM

          Subject: RE: [Behavioral-Finance] Dow Monthly returns

           

           

          Based upon the Dow’s price-only history as Mark has referenced below, when the year-to-date

          through August was down, as was the situation this year, September has been down about twice

          as often as it has been up.

          However, using total return (dividends reinvested) with our monthly CMI database for the 139

          years since 1870 (S&P 90 and 500 indexes starting in Jan ’26 and Jan ’50, respectively, and

          S&P-Wilshire 5000 since Apr ’01) – available upon request – there have been 45 instances with

          their years-to-date down through August, of which 28, or 62%, Septembers had negative returns.

          Furthermore, during the 12 second Presidential-term years since 1962, when the 48-month,

          socio-political-economic, or Kitchin, cycle fully emerged (see footnotes # 4, 10, 14 and 20

          here:  A Forecasting Model That Integrates Multiple Business and Stock Market Cycles

          September has been down 67% of the time, or twice as often as it has been up: eight times

          down compared to only four times up. 

          Considering only Septembers when both it was a Presidential term second year, typically the

          weakest of the four years (the political discord and economic uncertainty this year is unusually

          bearish) and the stock market, measured by our CMI, was down year-to-date through August,

          five of those seven instances were down or 71% of the time.

          Finally, four of those seven were during Supercycle Bear Market Periods (e.g., 1881-1896,

          1906-21, 1929-49, 1966-82 and 2000 to date) of which three, or 75%, of those Septembers

          had negative returns - 1966, 1974 and 2002 , while September 2006 gained 2.3%.  

          Since our work clearly shows that we’re still in a Supercycle Bear Market Period, in particular,

          the most devastating of the two kinds: a deflationary economic Supercycle Winter – see the

          summary of our quantified Supercycle Economic Seasons below - we believe the odds that

          this September, while it has started off with a strong start, will decline for the rest of the month,

          and it will probably even finish with a net negative return. 

          Of course, we have other reasons for expecting this also – see the summary of our forecasting

          models further below.

          Bob Bronson

          Bronson Capital Markets Research

          http://financialsense.com/editorials/bronson/main.html

          http://www.financialsense.com/contributors/bob-bronson

          http://www.financialsense.com/sites/default/files/users/u188/images/2010/supercycle-economic-seasons.jpg

          http://www.financialsense.com/sites/default/files/users/u188/images/2010/bronson-forecast-models.jpg

          -----Original Message-----
          From: Behavioral-Finance@yahoogroups.com [mailto:Behavioral-Finance@yahoogroups.com] On Behalf Of mark
          Sent: Sunday, September 05, 2010 1:49 PM
          To: Behavioral-Finance@yahoogroups.com
          Subject: [Behavioral-Finance] Dow Monthly returns

          This chart shows the Dow Industrials returns on a monthly basis back to 1886.  It is sorted on September returns

          http://www.mutualfundsystem.com/newsletters/DowMonthlyReturns.pdf

          ------------------------------------

          you may unsubscribe by sending an email to

          Behavioral-Finance-unsubscribe@yahoogroups.com

          Yahoo! Groups Links

          <*> To visit your group on the web, go to:

              http://groups.yahoo.com/group/Behavioral-Finance/

          <*> Your email settings:

              Individual Email | Traditional

          <*> To change settings online go to:

              http://groups.yahoo.com/group/Behavioral-Finance/join

              (Yahoo! ID required)

          <*> To change settings via email:

              Behavioral-Finance-digest@yahoogroups.com

              Behavioral-Finance-fullfeatured@yahoogroups.com

          <*> To unsubscribe from this group, send an email to:

              Behavioral-Finance-unsubscribe@yahoogroups.com

          <*> Your use of Yahoo! Groups is subject to:

              http://docs.yahoo.com/info/terms/

        • SergeyTS
          Annual only (=365 days). Economical cyles (the most important are Juglar and Kitchen) are here: http://www.timingsolution.com/TS/Forecast/2009_long_term/ They
          Message 4 of 15 , Sep 6, 2010
          View Source
          • 0 Attachment
            Annual only (=365 days).
             
            Economical cyles (the most important are Juglar and Kitchen) are here: http://www.timingsolution.com/TS/Forecast/2009_long_term/
             
            They are visible on this periodogram for Dow:
             
             
            Best regards.
            Sergey.
             
             
             
            ----- Original Message -----
            From: MarkK
            Sent: Monday, September 06, 2010 10:22 AM
            Subject: RE: [Behavioral-Finance] Dow Monthly returns

             

            Since there are numerous cycles involved what one are you using?
            Price? Economic? President? Elliot Wave? Hurts? And the list continues

            MarkK

            From: Behavioral-Finance@yahoogroups.com [mailto:Behavioral-Finance@yahoogroups.com] On Behalf Of SergeyTS
            Sent: Monday, September 06, 2010 9:49 AM
            To: Behavioral-Finance@yahoogroups.com
            Subject: Re: [Behavioral-Finance] Dow Monthly returns

             

            Hello,

            This is Annual cycle for Dow:

            the price history since 1789 year is used. Actually, Dow exists since 1885, before that the data are the interpolated prices of the most  liquid companies.

            Best regards.

            Sergey.

            ----- Original Message -----

            Sent: Monday, September 06, 2010 5:04 AM

            Subject: RE: [Behavioral-Finance] Dow Monthly returns

             

            Based upon the Dow’s price-only history as Mark has referenced below, when the year-to-date

            through August was down, as was the situation this year, September has been down about twice

            as often as it has been up.

            However, using total return (dividends reinvested) with our monthly CMI database for the 139

            years since 1870 (S&P 90 and 500 indexes starting in Jan ’26 and Jan ’50, respectively, and

            S&P-Wilshire 5000 since Apr ’01) – available upon request – there have been 45 instances with

            their years-to-date down through August, of which 28, or 62%, Septembers had negative returns.

            Furthermore, during the 12 second Presidential-term years since 1962, when the 48-month,

            socio-political-economic, or Kitchin, cycle fully emerged (see footnotes # 4, 10, 14 and 20

            here:  A Forecasting Model That Integrates Multiple Business and Stock Market Cycles

            September has been down 67% of the time, or twice as often as it has been up: eight times

            down compared to only four times up. 

            Considering only Septembers when both it was a Presidential term second year, typically the

            weakest of the four years (the political discord and economic uncertainty this year is unusually

            bearish) and the stock market, measured by our CMI, was down year-to-date through August,

            five of those seven instances were down or 71% of the time.

            Finally, four of those seven were during Supercycle Bear Market Periods (e.g., 1881-1896,

            1906-21, 1929-49, 1966-82 and 2000 to date) of which three, or 75%, of those Septembers

            had negative returns - 1966, 1974 and 2002 , while September 2006 gained 2.3%.  

            Since our work clearly shows that we’re still in a Supercycle Bear Market Period, in particular,

            the most devastating of the two kinds: a deflationary economic Supercycle Winter – see the

            summary of our quantified Supercycle Economic Seasons below - we believe the odds that

            this September, while it has started off with a strong start, will decline for the rest of the month,

            and it will probably even finish with a net negative return. 

            Of course, we have other reasons for expecting this also – see the summary of our forecasting

            models further below.

            Bob Bronson

            Bronson Capital Markets Research

            http://financialsense.com/editorials/bronson/main.html

            http://www.financialsense.com/contributors/bob-bronson

            http://www.financialsense.com/sites/default/files/users/u188/images/2010/supercycle-economic-seasons.jpg

            http://www.financialsense.com/sites/default/files/users/u188/images/2010/bronson-forecast-models.jpg

            -----Original Message-----
            From: Behavioral-Finance@yahoogroups.com [mailto:Behavioral-Finance@yahoogroups.com] On Behalf Of mark
            Sent: Sunday, September 05, 2010 1:49 PM
            To: Behavioral-Finance@yahoogroups.com
            Subject: [Behavioral-Finance] Dow Monthly returns

            This chart shows the Dow Industrials returns on a monthly basis back to 1886.  It is sorted on September returns

            http://www.mutualfundsystem.com/newsletters/DowMonthlyReturns.pdf

            ------------------------------------

            you may unsubscribe by sending an email to

            Behavioral-Finance-unsubscribe@yahoogroups.com

            Yahoo! Groups Links

            <*> To visit your group on the web, go to:

                http://groups.yahoo.com/group/Behavioral-Finance/

            <*> Your email settings:

                Individual Email | Traditional

            <*> To change settings online go to:

                http://groups.yahoo.com/group/Behavioral-Finance/join

                (Yahoo! ID required)

            <*> To change settings via email:

                Behavioral-Finance-digest@yahoogroups.com

                Behavioral-Finance-fullfeatured@yahoogroups.com

            <*> To unsubscribe from this group, send an email to:

                Behavioral-Finance-unsubscribe@yahoogroups.com

            <*> Your use of Yahoo! Groups is subject to:

                http://docs.yahoo.com/info/terms/

          • Bob Bronson
            Thanx, Sergey. Here are the charts for the calendar-year monthly pattern using price-only, total return (dividends reinvested) and real total return (dividends
            Message 5 of 15 , Sep 6, 2010
            View Source
            • 0 Attachment

              Thanx, Sergey. 

               

              Here are the charts for the calendar-year monthly pattern using price-only, total return (dividends reinvested)

              and real total return (dividends reinvested and adjusted for price inflation) using our CMI back to 1870.  Note

              that the three charts illustrate the average (mean) in red and the statistically more significant median in blue.

               

              The lower pair of red and blue lines in each chart is net effect of detrending the upper pair, which illustrates

              the monthly pattern after the bullish effect of the annualized return for the upper pair is removed.  In contrast

              to your detrended chart, which I discerned you presented from reading your very interesting website, that

              not only do the detrended pairs show that January has typically been higher that April, but that is especially

              so with the median indexes.

               

              However, your and our work agree that September has created the lowest intra-year month end on average,

              whether the stock market is measured with or without dividends and whether inflation is considered or not.

               

              In further follow up to my previous comments below, I was using five factors in my analysis: (a) stock

              market price; (b) (aggregated) dividends; (c) (price) inflation; (d) a simplistic form of price momentum,

              or actually stock market price velocity - year-to-date net price change through August; and (e) our notion

              of Supercycle Periods, which we quantify both fundamentally and technically, as was summarized there.

               

              (As most everyone here knows, physicists define momentum as mass times velocity, a vector, while most

              investors using price momentum are simply referring to a scalar metric, the rate of change of price, or its

              velocity: the first derivative, or first differences of discrete a price time series).)

               

              I say a simplistic form of momentum since as a scalar metric, I made my point about September’s history

              having been a down month using Bayesian if-then conditional logic, which quantifies independent variables

              (conditional probabilities) and dependent variables (unconditional probabilities) in a simplistic dichotomous

              manner of simply true or false, which in this case is monthly returns being up (positive returns) or down

              (negative returns).  The magnitude of those variables is entirely ignored. 

               

              If I presented a scattergram, or regression analysis, of simply using the monthly returns data as vector

              (with both plus or minus direction and the associated magnitude) a more quantitative case can be made

              about September’s likelihood of having a negative return since both the odds would be greater and they

              would be based on more than just the four instances I cited, wherein I introduced Supercycle Periods as

              a qualifying factor.  But in spite of the higher degrees of freedom, which some might wrong-headedly

              believe is (excess) data mining, of over fitting, we prefer the more qualitative approach as suggested

              by the confluence of confirming factors from indicators in our four-paired factors forecasting models

              than quantify monetary-economic, socio-political, valuation-sentiment and intra/inter market indicators.

               

              Finally, I note that if interest rates are added to the five factor analysis I introduced below, the probability

              for September having a negative return, and especially from now forward, would be even stronger,

              even just using Bayesian analysis of the historical data, since interest rates lead price inflation.  See

              those lead-lag comparison charts near the bottom of this article:

              Our August 10, 2010 Call: The Summer Stock Market Rally Is Over

              http://www.financialsense.com/contributors/bob-bronson/the-summer-stock-market-rally-is-over

               

               

               

              From: Behavioral-Finance@yahoogroups.com [mailto:Behavioral-Finance@yahoogroups.com] On Behalf Of SergeyTS
              Sent: Monday, September 06, 2010 7:49 AM
              To: Behavioral-Finance@yahoogroups.com
              Subject: Re: [Behavioral-Finance] Dow Monthly returns

               



              Hello,

               

              This is Annual cycle for Dow:

               

               

               

              the price history since 1789 year is used. Actually, Dow exists since 1885, before that the data are the interpolated prices of the most  liquid companies.

               

              Best regards.

              Sergey.

               

               

              ----- Original Message -----

              Sent: Monday, September 06, 2010 5:04 AM

              Subject: RE: [Behavioral-Finance] Dow Monthly returns

               

               

              Based upon the Dow’s price-only history as Mark has referenced below, when the year-to-date

              through August was down, as was the situation this year, September has been down about twice

              as often as it has been up.

              However, using total return (dividends reinvested) with our monthly CMI database for the 139

              years since 1870 (S&P 90 and 500 indexes starting in Jan ’26 and Jan ’50, respectively, and

              S&P-Wilshire 5000 since Apr ’01) – available upon request – there have been 45 instances with

              their years-to-date down through August, of which 28, or 62%, Septembers had negative returns.

              Furthermore, during the 12 second Presidential-term years since 1962, when the 48-month,

              socio-political-economic, or Kitchin, cycle fully emerged (see footnotes # 4, 10, 14 and 20

              here:  A Forecasting Model That Integrates Multiple Business and Stock Market Cycles

              September has been down 67% of the time, or twice as often as it has been up: eight times

              down compared to only four times up. 

              Considering only Septembers when both it was a Presidential term second year, typically the

              weakest of the four years (the political discord and economic uncertainty this year is unusually

              bearish) and the stock market, measured by our CMI, was down year-to-date through August,

              five of those seven instances were down or 71% of the time.

              Finally, four of those seven were during Supercycle Bear Market Periods (e.g., 1881-1896,

              1906-21, 1929-49, 1966-82 and 2000 to date) of which three, or 75%, of those Septembers

              had negative returns - 1966, 1974 and 2002 , while September 2006 gained 2.3%.  

              Since our work clearly shows that we’re still in a Supercycle Bear Market Period, in particular,

              the most devastating of the two kinds: a deflationary economic Supercycle Winter – see the

              summary of our quantified Supercycle Economic Seasons below - we believe the odds that

              this September, while it has started off with a strong start, will decline for the rest of the month,

              and it will probably even finish with a net negative return. 

              Of course, we have other reasons for expecting this also – see the summary of our forecasting

              models further below.

              Bob Bronson

              Bronson Capital Markets Research

              http://financialsense.com/editorials/bronson/main.html

              http://www.financialsense.com/contributors/bob-bronson

              http://www.financialsense.com/sites/default/files/users/u188/images/2010/supercycle-economic-seasons.jpg

              http://www.financialsense.com/sites/default/files/users/u188/images/2010/bronson-forecast-models.jpg

               

              -----Original Message-----
              From: Behavioral-Finance@yahoogroups.com [mailto:Behavioral-Finance@yahoogroups.com] On Behalf Of mark
              Sent: Sunday, September 05, 2010 1:49 PM
              To: Behavioral-Finance@yahoogroups.com
              Subject: [Behavioral-Finance] Dow Monthly returns

              This chart shows the Dow Industrials returns on a monthly basis back to 1886.  It is sorted on September returns

              http://www.mutualfundsystem.com/newsletters/DowMonthlyReturns.pdf

            • Bob Bronson
              Whoops, the referenced charts are now inserted below. From: Behavioral-Finance@yahoogroups.com [mailto:Behavioral-Finance@yahoogroups.com] On Behalf Of Bob
              Message 6 of 15 , Sep 6, 2010
              View Source
              • 0 Attachment

                Whoops, the referenced charts are now inserted below.

                 

                From: Behavioral-Finance@yahoogroups.com [mailto:Behavioral-Finance@yahoogroups.com] On Behalf Of Bob Bronson
                Sent: Monday, September 06, 2010 4:53 PM
                To: Behavioral-Finance@yahoogroups.com
                Subject: RE: [Behavioral-Finance] Dow Monthly returns


                Thanx, Sergey. 

                 

                Here are the charts for the calendar-year monthly pattern using price-only, total return (dividends reinvested)

                and real total return (dividends reinvested and adjusted for price inflation) using our CMI back to 1870.  Note

                that the three charts illustrate the average (mean) in red and the statistically more significant median in blue.

                 

                The lower pair of red and blue lines in each chart is net effect of detrending the upper pair, which illustrates

                the monthly pattern after the bullish effect of the annualized return for the upper pair is removed.  In contrast

                to your detrended chart, which I discerned you presented from reading your very interesting website, that

                not only do the detrended pairs show that January has typically been higher that April, but that is especially

                so with the median indexes.

                 

                However, your and our work agree that September has created the lowest intra-year month end on average,

                whether the stock market is measured with or without dividends and whether inflation is considered or not.

                 

                In further follow up to my previous comments below, I was using five factors in my analysis: (a) stock

                market price; (b) (aggregated) dividends; (c) (price) inflation; (d) a simplistic form of price momentum,

                or actually stock market price velocity - year-to-date net price change through August; and (e) our notion

                of Supercycle Periods, which we quantify both fundamentally and technically, as was summarized there.

                 

                (As most everyone here knows, physicists define momentum as mass times velocity, a vector, while most

                investors using price momentum are simply referring to a scalar metric, the rate of change of price, or its

                velocity: the first derivative, or first differences of discrete a price time series).)

                 

                I say a simplistic form of momentum since as a scalar metric, I made my point about September’s history

                having been a down month using Bayesian if-then conditional logic, which quantifies independent variables

                (conditional probabilities) and dependent variables (unconditional probabilities) in a simplistic dichotomous

                manner of simply true or false, which in this case is monthly returns being up (positive returns) or down

                (negative returns).  The magnitude of those variables is entirely ignored. 

                 

                If I presented a scattergram, or regression analysis, of simply using the monthly returns data as vector

                (with both plus or minus direction and the associated magnitude) a more quantitative case can be made

                about September’s likelihood of having a negative return since both the odds would be greater and they

                would be based on more than just the four instances I cited, wherein I introduced Supercycle Periods as

                a qualifying factor.  But in spite of the higher degrees of freedom, which some might wrong-headedly

                believe is (excess) data mining, of over fitting, we prefer the more qualitative approach as suggested

                by the confluence of confirming factors from indicators in our four-paired factors forecasting models

                than quantify monetary-economic, socio-political, valuation-sentiment and intra/inter market indicators.

                 

                Finally, I note that if interest rates are added to the five factor analysis I introduced below, the probability

                for September having a negative return, and especially from now forward, would be even stronger,

                even just using Bayesian analysis of the historical data, since interest rates lead price inflation.  See

                those lead-lag comparison charts near the bottom of this article:

                Our August 10, 2010 Call: The Summer Stock Market Rally Is Over

                http://www.financialsense.com/contributors/bob-bronson/the-summer-stock-market-rally-is-over

                 

                 

                From: Behavioral-Finance@yahoogroups.com [mailto:Behavioral-Finance@yahoogroups.com] On Behalf Of SergeyTS
                Sent: Monday, September 06, 2010 7:49 AM
                To: Behavioral-Finance@yahoogroups.com
                Subject: Re: [Behavioral-Finance] Dow Monthly returns

                 

                 

                Hello,

                 

                This is Annual cycle for Dow:

                 

                 

                 

                the price history since 1789 year is used. Actually, Dow exists since 1885, before that the data are the interpolated prices of the most  liquid companies.

                 

                Best regards.

                Sergey.

                 

                 

                ----- Original Message -----

                Sent: Monday, September 06, 2010 5:04 AM

                Subject: RE: [Behavioral-Finance] Dow Monthly returns

                 

                 

                Based upon the Dow’s price-only history as Mark has referenced below, when the year-to-date

                through August was down, as was the situation this year, September has been down about twice

                as often as it has been up.

                However, using total return (dividends reinvested) with our monthly CMI database for the 139

                years since 1870 (S&P 90 and 500 indexes starting in Jan ’26 and Jan ’50, respectively, and

                S&P-Wilshire 5000 since Apr ’01) – available upon request – there have been 45 instances with

                their years-to-date down through August, of which 28, or 62%, Septembers had negative returns.

                Furthermore, during the 12 second Presidential-term years since 1962, when the 48-month,

                socio-political-economic, or Kitchin, cycle fully emerged (see footnotes # 4, 10, 14 and 20

                here:  A Forecasting Model That Integrates Multiple Business and Stock Market Cycles

                September has been down 67% of the time, or twice as often as it has been up: eight times

                down compared to only four times up. 

                Considering only Septembers when both it was a Presidential term second year, typically the

                weakest of the four years (the political discord and economic uncertainty this year is unusually

                bearish) and the stock market, measured by our CMI, was down year-to-date through August,

                five of those seven instances were down or 71% of the time.

                Finally, four of those seven were during Supercycle Bear Market Periods (e.g., 1881-1896,

                1906-21, 1929-49, 1966-82 and 2000 to date) of which three, or 75%, of those Septembers

                had negative returns - 1966, 1974 and 2002 , while September 2006 gained 2.3%.  

                Since our work clearly shows that we’re still in a Supercycle Bear Market Period, in particular,

                the most devastating of the two kinds: a deflationary economic Supercycle Winter – see the

                summary of our quantified Supercycle Economic Seasons below - we believe the odds that

                this September, while it has started off with a strong start, will decline for the rest of the month,

                and it will probably even finish with a net negative return. 

                Of course, we have other reasons for expecting this also – see the summary of our forecasting

                models further below.

                Bob Bronson

                Bronson Capital Markets Research

                http://financialsense.com/editorials/bronson/main.html

                http://www.financialsense.com/contributors/bob-bronson

                http://www.financialsense.com/sites/default/files/users/u188/images/2010/supercycle-economic-seasons.jpg

                http://www.financialsense.com/sites/default/files/users/u188/images/2010/bronson-forecast-models.jpg

                 

                -----Original Message-----
                From: Behavioral-Finance@yahoogroups.com [mailto:Behavioral-Finance@yahoogroups.com] On Behalf Of mark
                Sent: Sunday, September 05, 2010 1:49 PM
                To: Behavioral-Finance@yahoogroups.com
                Subject: [Behavioral-Finance] Dow Monthly returns

                This chart shows the Dow Industrials returns on a monthly basis back to 1886.  It is sorted on September returns

                http://www.mutualfundsystem.com/newsletters/DowMonthlyReturns.pdf

                 




              • pgreenfinch
                Hello, everybody ! Bingo, this debate, however interesting it is, confirms my initial remark, which has still not been answered, about the relations (or lack
                Message 7 of 15 , Sep 7, 2010
                View Source
                • 0 Attachment
                  Hello, everybody !
                   
                  Bingo, this debate, however interesting it is, confirms my initial
                  remark, which has still not been answered, about the relations  
                  (or lack of them) between BF and TA (or the other way round),
                  as there is not a word of explanation about the individual and
                  collective behaviors behind those data and those analyses
                   
                  The zigzags are here, the metrics are here, the "join the dot"
                  job is made (the representative heuristic as Martin says),
                  efficient conclusion might be or not drawn for trading and
                  investing, there is some magic shown in those nearly cosmic
                  cycles, but no explanation is given about the influence of
                  those sexy curves on the players' testosterone ;-).
                   
                  Where is the beef ? Oops, where is the BF ;) ?
                   
                  Peter
                   
                   
                  ----- Original Message -----
                  Sent: Tuesday, September 07, 2010 1:23 AM
                  Subject: [Norton AntiSpam] RE: [Behavioral-Finance] Dow Monthly returns

                   

                  Whoops, the referenced charts are now inserted below.

                  From: Behavioral-Finance@yahoogroups.com [mailto:Behavioral-Finance@yahoogroups.com] On Behalf Of Bob Bronson
                  Sent: Monday, September 06, 2010 4:53 PM
                  To: Behavioral-Finance@yahoogroups.com
                  Subject: RE: [Behavioral-Finance] Dow Monthly returns


                  Thanx, Sergey. 

                  Here are the charts for the calendar-year monthly pattern using price-only, total return (dividends reinvested)

                  and real total return (dividends reinvested and adjusted for price inflation) using our CMI back to 1870.  Note

                  that the three charts illustrate the average (mean) in red and the statistically more significant median in blue.

                  The lower pair of red and blue lines in each chart is net effect of detrending the upper pair, which illustrates

                  the monthly pattern after the bullish effect of the annualized return for the upper pair is removed.  In contrast

                  to your detrended chart, which I discerned you presented from reading your very interesting website, that

                  not only do the detrended pairs show that January has typically been higher that April, but that is especially

                  so with the median indexes.

                  However, your and our work agree that September has created the lowest intra-year month end on average,

                  whether the stock market is measured with or without dividends and whether inflation is considered or not.

                  In further follow up to my previous comments below, I was using five factors in my analysis: (a) stock

                  market price; (b) (aggregated) dividends; (c) (price) inflation; (d) a simplistic form of price momentum,

                  or actually stock market price velocity - year-to-date net price change through August; and (e) our notion

                  of Supercycle Periods, which we quantify both fundamentally and technically, as was summarized there.

                  (As most everyone here knows, physicists define momentum as mass times velocity, a vector, while most

                  investors using price momentum are simply referring to a scalar metric, the rate of change of price, or its

                  velocity: the first derivative, or first differences of discrete a price time series).)

                  I say a simplistic form of momentum since as a scalar metric, I made my point about September’s history

                  having been a down month using Bayesian if-then conditional logic, which quantifies independent variables

                  (conditional probabilities) and dependent variables (unconditional probabilities) in a simplistic dichotomous

                  manner of simply true or false, which in this case is monthly returns being up (positive returns) or down

                  (negative returns).  The magnitude of those variables is entirely ignored. 

                  If I presented a scattergram, or regression analysis, of simply using the monthly returns data as vector

                  (with both plus or minus direction and the associated magnitude) a more quantitative case can be made

                  about September’s likelihood of having a negative return since both the odds would be greater and they

                  would be based on more than just the four instances I cited, wherein I introduced Supercycle Periods as

                  a qualifying factor.  But in spite of the higher degrees of freedom, which some might wrong-headedly

                  believe is (excess) data mining, of over fitting, we prefer the more qualitative approach as suggested

                  by the confluence of confirming factors from indicators in our four-paired factors forecasting models

                  than quantify monetary-economic, socio-political, valuation-sentiment and intra/inter market indicators.

                  Finally, I note that if interest rates are added to the five factor analysis I introduced below, the probability

                  for September having a negative return, and especially from now forward, would be even stronger,

                  even just using Bayesian analysis of the historical data, since interest rates lead price inflation.  See

                  those lead-lag comparison charts near the bottom of this article:

                  Our August 10, 2010 Call: The Summer Stock Market Rally Is Over

                  http://www.financialsense.com/contributors/bob-bronson/the-summer-stock-market-rally-is-over

                  From: Behavioral-Finance@yahoogroups.com [mailto:Behavioral-Finance@yahoogroups.com] On Behalf Of SergeyTS
                  Sent: Monday, September 06, 2010 7:49 AM
                  To: Behavioral-Finance@yahoogroups.com
                  Subject: Re: [Behavioral-Finance] Dow Monthly returns

                  Hello,

                  This is Annual cycle for Dow:

                  the price history since 1789 year is used. Actually, Dow exists since 1885, before that the data are the interpolated prices of the most  liquid companies.

                  Best regards.

                  Sergey.

                  ----- Original Message -----

                  Sent: Monday, September 06, 2010 5:04 AM

                  Subject: RE: [Behavioral-Finance] Dow Monthly returns

                   

                  Based upon the Dow’s price-only history as Mark has referenced below, when the year-to-date

                  through August was down, as was the situation this year, September has been down about twice

                  as often as it has been up.

                  However, using total return (dividends reinvested) with our monthly CMI database for the 139

                  years since 1870 (S&P 90 and 500 indexes starting in Jan ’26 and Jan ’50, respectively, and

                  S&P-Wilshire 5000 since Apr ’01) – available upon request – there have been 45 instances with

                  their years-to-date down through August, of which 28, or 62%, Septembers had negative returns.

                  Furthermore, during the 12 second Presidential-term years since 1962, when the 48-month,

                  socio-political-economic, or Kitchin, cycle fully emerged (see footnotes # 4, 10, 14 and 20

                  here:  A Forecasting Model That Integrates Multiple Business and Stock Market Cycles

                  September has been down 67% of the time, or twice as often as it has been up: eight times

                  down compared to only four times up. 

                  Considering only Septembers when both it was a Presidential term second year, typically the

                  weakest of the four years (the political discord and economic uncertainty this year is unusually

                  bearish) and the stock market, measured by our CMI, was down year-to-date through August,

                  five of those seven instances were down or 71% of the time.

                  Finally, four of those seven were during Supercycle Bear Market Periods (e.g., 1881-1896,

                  1906-21, 1929-49, 1966-82 and 2000 to date) of which three, or 75%, of those Septembers

                  had negative returns - 1966, 1974 and 2002 , while September 2006 gained 2.3%.  

                  Since our work clearly shows that we’re still in a Supercycle Bear Market Period, in particular,

                  the most devastating of the two kinds: a deflationary economic Supercycle Winter – see the

                  summary of our quantified Supercycle Economic Seasons below - we believe the odds that

                  this September, while it has started off with a strong start, will decline for the rest of the month,

                  and it will probably even finish with a net negative return. 

                  Of course, we have other reasons for expecting this also – see the summary of our forecasting

                  models further below.

                  Bob Bronson

                  Bronson Capital Markets Research

                  http://financialsense.com/editorials/bronson/main.html

                  http://www.financialsense.com/contributors/bob-bronson

                  http://www.financialsense.com/sites/default/files/users/u188/images/2010/supercycle-economic-seasons.jpg

                  http://www.financialsense.com/sites/default/files/users/u188/images/2010/bronson-forecast-models.jpg

                  -----Original Message-----
                  From: Behavioral-Finance@yahoogroups.com [mailto:Behavioral-Finance@yahoogroups.com] On Behalf Of mark
                  Sent: Sunday, September 05, 2010 1:49 PM
                  To: Behavioral-Finance@yahoogroups.com
                  Subject: [Behavioral-Finance] Dow Monthly returns

                  This chart shows the Dow Industrials returns on a monthly basis back to 1886.  It is sorted on September returns

                  http://www.mutualfundsystem.com/newsletters/DowMonthlyReturns.pdf




                • SergeyTS
                  Hello, Bob: Thank you, this is a very interesting information. I want to add just a few words re Annual cycle. Actually it has not many degrees of freedom, and
                  Message 8 of 15 , Sep 7, 2010
                  View Source
                  • 0 Attachment
                    Hello, Bob:
                     
                    Thank you, this is a very interesting information.
                     
                    I want to add just a few words re Annual cycle. Actually it has not many degrees of freedom, and the overfitting there is not possible. This is a very simple model.
                     
                    But - I would not go there now. What I am more interested to discuss in regards to Technical Analysis is other things.My point is that the stock market is changing its structure with the time. If we look at any indicator/factor within some significant period of time, we see that. The Annual cycle is just one way to see that. Consider the market and the Annual cycle now and 100 years ago. The Annual cycle was stronger/more obvious 100 years ago than now. (And there are fundamentals involved: agricultural sector now is 2% while then it was 40-50%  of GDP, something like that, I do not remember exact numbers). Also, I have observed that the Annual cycle becomes stronger (or, in other words, its role becomes more obvious) at crisis times. Maybe during crisis we all tend to go back to old things that worked for us in the apst, I do not know.
                     
                    Please look at this diagram, it shows the level of activity of the Annual cycle within the latest 120 years. You can see there that during the crisises of 1970s, in 1987, 2001 the Annual cycle was more active (more red on the diagram):
                     
                     
                    Best regards,
                     
                    Sergey
                     
                     
                    ----- Original Message -----
                    Sent: Monday, September 06, 2010 7:23 PM
                    Subject: RE: [Behavioral-Finance] Dow Monthly returns

                     

                    Whoops, the referenced charts are now inserted below.

                    From: Behavioral-Finance@yahoogroups.com [mailto:Behavioral-Finance@yahoogroups.com] On Behalf Of Bob Bronson
                    Sent: Monday, September 06, 2010 4:53 PM
                    To: Behavioral-Finance@yahoogroups.com
                    Subject: RE: [Behavioral-Finance] Dow Monthly returns


                    Thanx, Sergey. 

                    Here are the charts for the calendar-year monthly pattern using price-only, total return (dividends reinvested)

                    and real total return (dividends reinvested and adjusted for price inflation) using our CMI back to 1870.  Note

                    that the three charts illustrate the average (mean) in red and the statistically more significant median in blue.

                    The lower pair of red and blue lines in each chart is net effect of detrending the upper pair, which illustrates

                    the monthly pattern after the bullish effect of the annualized return for the upper pair is removed.  In contrast

                    to your detrended chart, which I discerned you presented from reading your very interesting website, that

                    not only do the detrended pairs show that January has typically been higher that April, but that is especially

                    so with the median indexes.

                    However, your and our work agree that September has created the lowest intra-year month end on average,

                    whether the stock market is measured with or without dividends and whether inflation is considered or not.

                    In further follow up to my previous comments below, I was using five factors in my analysis: (a) stock

                    market price; (b) (aggregated) dividends; (c) (price) inflation; (d) a simplistic form of price momentum,

                    or actually stock market price velocity - year-to-date net price change through August; and (e) our notion

                    of Supercycle Periods, which we quantify both fundamentally and technically, as was summarized there.

                    (As most everyone here knows, physicists define momentum as mass times velocity, a vector, while most

                    investors using price momentum are simply referring to a scalar metric, the rate of change of price, or its

                    velocity: the first derivative, or first differences of discrete a price time series).)

                    I say a simplistic form of momentum since as a scalar metric, I made my point about September’s history

                    having been a down month using Bayesian if-then conditional logic, which quantifies independent variables

                    (conditional probabilities) and dependent variables (unconditional probabilities) in a simplistic dichotomous

                    manner of simply true or false, which in this case is monthly returns being up (positive returns) or down

                    (negative returns).  The magnitude of those variables is entirely ignored. 

                    If I presented a scattergram, or regression analysis, of simply using the monthly returns data as vector

                    (with both plus or minus direction and the associated magnitude) a more quantitative case can be made

                    about September’s likelihood of having a negative return since both the odds would be greater and they

                    would be based on more than just the four instances I cited, wherein I introduced Supercycle Periods as

                    a qualifying factor.  But in spite of the higher degrees of freedom, which some might wrong-headedly

                    believe is (excess) data mining, of over fitting, we prefer the more qualitative approach as suggested

                    by the confluence of confirming factors from indicators in our four-paired factors forecasting models

                    than quantify monetary-economic, socio-political, valuation-sentiment and intra/inter market indicators.

                    Finally, I note that if interest rates are added to the five factor analysis I introduced below, the probability

                    for September having a negative return, and especially from now forward, would be even stronger,

                    even just using Bayesian analysis of the historical data, since interest rates lead price inflation.  See

                    those lead-lag comparison charts near the bottom of this article:

                    Our August 10, 2010 Call: The Summer Stock Market Rally Is Over

                    http://www.financialsense.com/contributors/bob-bronson/the-summer-stock-market-rally-is-over

                    From: Behavioral-Finance@yahoogroups.com [mailto:Behavioral-Finance@yahoogroups.com] On Behalf Of SergeyTS
                    Sent: Monday, September 06, 2010 7:49 AM
                    To: Behavioral-Finance@yahoogroups.com
                    Subject: Re: [Behavioral-Finance] Dow Monthly returns

                    Hello,

                    This is Annual cycle for Dow:

                    the price history since 1789 year is used. Actually, Dow exists since 1885, before that the data are the interpolated prices of the most  liquid companies.

                    Best regards.

                    Sergey.

                    ----- Original Message -----

                    Sent: Monday, September 06, 2010 5:04 AM

                    Subject: RE: [Behavioral-Finance] Dow Monthly returns

                     

                    Based upon the Dow’s price-only history as Mark has referenced below, when the year-to-date

                    through August was down, as was the situation this year, September has been down about twice

                    as often as it has been up.

                    However, using total return (dividends reinvested) with our monthly CMI database for the 139

                    years since 1870 (S&P 90 and 500 indexes starting in Jan ’26 and Jan ’50, respectively, and

                    S&P-Wilshire 5000 since Apr ’01) – available upon request – there have been 45 instances with

                    their years-to-date down through August, of which 28, or 62%, Septembers had negative returns.

                    Furthermore, during the 12 second Presidential-term years since 1962, when the 48-month,

                    socio-political-economic, or Kitchin, cycle fully emerged (see footnotes # 4, 10, 14 and 20

                    here:  A Forecasting Model That Integrates Multiple Business and Stock Market Cycles

                    September has been down 67% of the time, or twice as often as it has been up: eight times

                    down compared to only four times up. 

                    Considering only Septembers when both it was a Presidential term second year, typically the

                    weakest of the four years (the political discord and economic uncertainty this year is unusually

                    bearish) and the stock market, measured by our CMI, was down year-to-date through August,

                    five of those seven instances were down or 71% of the time.

                    Finally, four of those seven were during Supercycle Bear Market Periods (e.g., 1881-1896,

                    1906-21, 1929-49, 1966-82 and 2000 to date) of which three, or 75%, of those Septembers

                    had negative returns - 1966, 1974 and 2002 , while September 2006 gained 2.3%.  

                    Since our work clearly shows that we’re still in a Supercycle Bear Market Period, in particular,

                    the most devastating of the two kinds: a deflationary economic Supercycle Winter – see the

                    summary of our quantified Supercycle Economic Seasons below - we believe the odds that

                    this September, while it has started off with a strong start, will decline for the rest of the month,

                    and it will probably even finish with a net negative return. 

                    Of course, we have other reasons for expecting this also – see the summary of our forecasting

                    models further below.

                    Bob Bronson

                    Bronson Capital Markets Research

                    http://financialsense.com/editorials/bronson/main.html

                    http://www.financialsense.com/contributors/bob-bronson

                    http://www.financialsense.com/sites/default/files/users/u188/images/2010/supercycle-economic-seasons.jpg

                    http://www.financialsense.com/sites/default/files/users/u188/images/2010/bronson-forecast-models.jpg

                    -----Original Message-----
                    From: Behavioral-Finance@yahoogroups.com [mailto:Behavioral-Finance@yahoogroups.com] On Behalf Of mark
                    Sent: Sunday, September 05, 2010 1:49 PM
                    To: Behavioral-Finance@yahoogroups.com
                    Subject: [Behavioral-Finance] Dow Monthly returns

                    This chart shows the Dow Industrials returns on a monthly basis back to 1886.  It is sorted on September returns

                    http://www.mutualfundsystem.com/newsletters/DowMonthlyReturns.pdf




                  • Bob Bronson
                    Sergey, I agree there are not too many degrees of (statistical) freedom in ferreting out the monthly pattern of the Annual Cycle. My point there was in
                    Message 9 of 15 , Sep 7, 2010
                    View Source
                    • 0 Attachment

                      Sergey,

                       

                      I agree there are not too many degrees of (statistical) freedom in ferreting out

                      the monthly pattern of the Annual Cycle.  My point there was in reference to

                      using five factors, or conditional filters, to make my point about this September

                      most likely being a down month in spite of its first three day’s strong start. 

                       

                      Furthermore, it would not be surprising if Friday’s stock market intraday high

                      was/is the high for the month, especially if tomorrow’s intraday action doesn’t

                      make a higher high.  But if does, it probably will be only a slightly higher high

                      and then it will probably net decline over the rest of the month.  By the way,

                      on a daily rather than only month-end pricing basis, October 7 is the historical

                      average low point, especially at the end of the Four Year Cycle.

                       

                      As far as the stock market changing its structure over time, consider that some

                      persistent anomalies, or non-random patterns, like the monthly pattern of the

                      annual cycle, may be robust, but aperiodically vary because of market efficiency,

                      or what we call competitive participant-observer feedback (gaming) on such

                      everybody-knows patterns.  We have noted before that the six-to-seven month

                      Strong Seasons and five-to-six month Weak Seasons of the Annual Cycle (not

                      calendar) are very robust because they are fundamentally grounded and are

                      not likely to disappear, at least due to market efficiency, but they certainly

                      exhibit pattern alternation (recall the Oct 2007 high and the Mar 2009 low).

                       

                      We’ve also pointed out to those on our private email list that some shorter term

                      calendar patterns, or time cycles, like the Yearend Holiday Season patterns,

                      which has been socialized into existence, are so gamed that they pattern

                      alternate being bigger or smaller than their historical average.  In this way

                      they still can maintain their historical average on an ex post basis, but vary

                      from it sufficiently to be non-profitable for those who simply play the average

                      pattern.  While, I believe this partially explains your observation of the varying

                      intensity of the Annual Cycle illustrated in your chart below, but I, for one,

                      would like to know more about your crisis theory of the variation in their activity.

                       

                      Conceptually and empirically, even institutionalized (e.g., quarterly financial

                      reporting) and socialized (e.g. holiday) time cycles can become somewhat of

                      a non-occurrence because of such feedback-driven pattern alteration, in which

                      case we say they exhibit “meaningless means.”  This apparent oddity is kind

                      of like the center of gravity being an out-of-body point like with a hula hoop

                      or a cantilever.  Similarly, and despite their being common in mathematics,

                      asymptotic patterns are extremely rare in the capital markets, so we prefer

                      to call the most powerful pricing force reversion-to-the-extreme behavior.

                       

                      In other words, the market’s efficiency does not require that a non-random

                      pattern’s average performance completely disappear, but rather it will become

                      predictably non-profitable, even if it is fundamentally grounded like the Strong

                      and Weak Seasons of the Annual Cycle, the Yearend Holiday Season and

                      the Four Year Cycle (recall the year late occurrence in 1986-7 and the cycle

                      inversion in 2006).

                       

                      Pattern alternation is one of several reasons why we believe the Semi-Weak

                      Case of the Efficient Market Hypothesis is the most operative, as I’ve pointed

                      out before in this forum when I noted there are consistent A players, like in

                      most any game as trading/investing surely is, despite random-walk arguments

                      to the contrary. 

                       

                      In this regard, I will be soon sending to our private email list (which anyone

                      here can join) a report, Moody Market, that presents heretofore unpublished

                      (as far as we know) aspects of the distribution of price returns over many time

                      frames that are very interesting.  They can be explained by the trend-following

                      (herding) behavior of institutional and individual investors and traders.

                       

                       

                       

                      _________________________________________________________________

                      From: Behavioral-Finance@yahoogroups.com [mailto:Behavioral-Finance@yahoogroups.com]

                      On Behalf Of SergeyTS
                      Sent: Tuesday, September 07, 2010 6:21 PM
                      To: Behavioral-Finance@yahoogroups.com
                      Subject: Re: [Behavioral-Finance] Dow Monthly returns

                       

                       

                      Hello, Bob:

                       

                      Thank you, this is a very interesting information.

                       

                      I want to add just a few words re Annual cycle. Actually it has not many degrees of freedom, and the overfitting there is not possible. This is a very simple model.

                       

                      But - I would not go there now. What I am more interested to discuss in regards to Technical Analysis is other things.My point is that the stock market is changing its structure with the time. If we look at any indicator/factor within some significant period of time, we see that. The Annual cycle is just one way to see that. Consider the market and the Annual cycle now and 100 years ago. The Annual cycle was stronger/more obvious 100 years ago than now. (And there are fundamentals involved: agricultural sector now is 2% while then it was 40-50%  of GDP, something like that, I do not remember exact numbers). Also, I have observed that the Annual cycle becomes stronger (or, in other words, its role becomes more obvious) at crisis times. Maybe during crisis we all tend to go back to old things that worked for us in the apst, I do not know.

                       

                      Please look at this diagram, it shows the level of activity of the Annual cycle within the latest 120 years. You can see there that during the crisises of 1970s, in 1987, 2001 the Annual cycle was more active (more red on the diagram):

                       

                       

                      Best regards,

                       

                      Sergey

                       

                       

                      ----- Original Message -----

                      Sent: Monday, September 06, 2010 7:23 PM

                      Subject: RE: [Behavioral-Finance] Dow Monthly returns

                       

                       

                      Whoops, the referenced charts are now inserted below.

                      From: Behavioral-Finance@yahoogroups.com [mailto:Behavioral-Finance@yahoogroups.com] On Behalf Of Bob Bronson
                      Sent: Monday, September 06, 2010 4:53 PM
                      To: Behavioral-Finance@yahoogroups.com
                      Subject: RE: [Behavioral-Finance] Dow Monthly returns



                      Thanx, Sergey. 

                      Here are the charts for the calendar-year monthly pattern using price-only, total return (dividends reinvested)

                      and real total return (dividends reinvested and adjusted for price inflation) using our CMI back to 1870.  Note

                      that the three charts illustrate the average (mean) in red and the statistically more significant median in blue.

                      The lower pair of red and blue lines in each chart is net effect of detrending the upper pair, which illustrates

                      the monthly pattern after the bullish effect of the annualized return for the upper pair is removed.  In contrast

                      to your detrended chart, which I discerned you presented from reading your very interesting website, that

                      not only do the detrended pairs show that January has typically been higher that April, but that is especially

                      so with the median indexes.

                      However, your and our work agree that September has created the lowest intra-year month end on average,

                      whether the stock market is measured with or without dividends and whether inflation is considered or not.

                      In further follow up to my previous comments below, I was using five factors in my analysis: (a) stock

                      market price; (b) (aggregated) dividends; (c) (price) inflation; (d) a simplistic form of price momentum,

                      or actually stock market price velocity - year-to-date net price change through August; and (e) our notion

                      of Supercycle Periods, which we quantify both fundamentally and technically, as was summarized there.

                      (As most everyone here knows, physicists define momentum as mass times velocity, a vector, while most

                      investors using price momentum are simply referring to a scalar metric, the rate of change of price, or its

                      velocity: the first derivative, or first differences of discrete a price time series).)

                      I say a simplistic form of momentum since as a scalar metric, I made my point about September’s history

                      having been a down month using Bayesian if-then conditional logic, which quantifies independent variables

                      (conditional probabilities) and dependent variables (unconditional probabilities) in a simplistic dichotomous

                      manner of simply true or false, which in this case is monthly returns being up (positive returns) or down

                      (negative returns).  The magnitude of those variables is entirely ignored. 

                      If I presented a scattergram, or regression analysis, of simply using the monthly returns data as vector

                      (with both plus or minus direction and the associated magnitude) a more quantitative case can be made

                      about September’s likelihood of having a negative return since both the odds would be greater and they

                      would be based on more than just the four instances I cited, wherein I introduced Supercycle Periods as

                      a qualifying factor.  But in spite of the higher degrees of freedom, which some might wrong-headedly

                      believe is (excess) data mining, of over fitting, we prefer the more qualitative approach as suggested

                      by the confluence of confirming factors from indicators in our four-paired factors forecasting models

                      that quantify monetary-economic, socio-political, valuation-sentiment and intra/inter market indicators.

                      Finally, I note that if interest rates are added to the five factor analysis I introduced below, the probability

                      for September having a negative return, and especially from now forward, would be even stronger,

                      even just using Bayesian analysis of the historical data since interest rates lead price inflation.  See

                      those lead-lag comparison charts near the bottom of this article:

                      Our August 10, 2010 Call: The Summer Stock Market Rally Is Over

                      http://www.financialsense.com/contributors/bob-bronson/the-summer-stock-market-rally-is-over

                      From: Behavioral-Finance@yahoogroups.com [mailto:Behavioral-Finance@yahoogroups.com] On Behalf Of SergeyTS
                      Sent: Monday, September 06, 2010 7:49 AM
                      To: Behavioral-Finance@yahoogroups.com
                      Subject: Re: [Behavioral-Finance] Dow Monthly returns

                      Hello,

                      This is Annual cycle for Dow:

                      the price history since 1789 year is used. Actually, Dow exists since 1885, before that the data are the interpolated prices of the most  liquid companies.

                      Best regards.

                      Sergey.

                      ----- Original Message -----

                      Sent: Monday, September 06, 2010 5:04 AM

                      Subject: RE: [Behavioral-Finance] Dow Monthly returns

                       

                      Based upon the Dow’s price-only history as Mark has referenced below, when the year-to-date

                      through August was down, as was the situation this year, September has been down about twice

                      as often as it has been up.

                      However, using total return (dividends reinvested) with our monthly CMI database for the 139

                      years since 1870 (S&P 90 and 500 indexes starting in Jan ’26 and Jan ’50, respectively, and

                      S&P-Wilshire 5000 since Apr ’01) – available upon request – there have been 45 instances with

                      their years-to-date down through August, of which 28, or 62%, Septembers had negative returns.

                      Furthermore, during the 12 second Presidential-term years since 1962, when the 48-month,

                      socio-political-economic, or Kitchin, cycle fully emerged (see footnotes # 4, 10, 14 and 20

                      here:  A Forecasting Model That Integrates Multiple Business and Stock Market Cycles

                      September has been down 67% of the time, or twice as often as it has been up: eight times

                      down compared to only four times up. 

                      Considering only Septembers when both it was a Presidential term second year, typically the

                      weakest of the four years (the political discord and economic uncertainty this year is unusually

                      bearish) and the stock market, measured by our CMI, was down year-to-date through August,

                      five of those seven instances were down or 71% of the time.

                      Finally, four of those seven were during Supercycle Bear Market Periods (e.g., 1881-1896,

                      1906-21, 1929-49, 1966-82 and 2000 to date) of which three, or 75%, of those Septembers

                      had negative returns - 1966, 1974 and 2002 , while September 2006 gained 2.3%.  

                      Since our work clearly shows that we’re still in a Supercycle Bear Market Period, in particular,

                      the most devastating of the two kinds: a deflationary economic Supercycle Winter – see the

                      summary of our quantified Supercycle Economic Seasons below - we believe the odds that

                      this September, while it has started off with a strong start, will decline for the rest of the month,

                      and it will probably even finish with a net negative return. 

                      Of course, we have other reasons for expecting this also – see the summary of our forecasting

                      models further below.

                      Bob Bronson

                      Bronson Capital Markets Research

                      http://financialsense.com/editorials/bronson/main.html

                      http://www.financialsense.com/contributors/bob-bronson

                      http://www.financialsense.com/sites/default/files/users/u188/images/2010/supercycle-economic-seasons.jpg

                      http://www.financialsense.com/sites/default/files/users/u188/images/2010/bronson-forecast-models.jpg

                      -----Original Message-----
                      From: Behavioral-Finance@yahoogroups.com [mailto:Behavioral-Finance@yahoogroups.com] On Behalf Of mark
                      Sent: Sunday, September 05, 2010 1:49 PM
                      To: Behavioral-Finance@yahoogroups.com
                      Subject: [Behavioral-Finance] Dow Monthly returns

                      This chart shows the Dow Industrials returns on a monthly basis back to 1886.  It is sorted on September returns

                      http://www.mutualfundsystem.com/newsletters/DowMonthlyReturns.pdf








                    • Jerry Wagner
                      Bob and Sergey Below is a chart (the blue line) we publish in January of each year - I ve been doing so for about 20 years. Each week we chart the actual
                      Message 10 of 15 , Sep 8, 2010
                      View Source
                      • 0 Attachment

                        Bob and Sergey

                         

                        Below is a chart (the blue line) we publish in January of each year – I’ve been doing so for about 20 years.  Each week we chart the actual performance of the Dow (the red line) versus the projection.  We also have a strategy that trades its buy and sell points (which are available a year in advance).  Uncanny how accurate it can be at times (especially spring and fall).  Last week, for example, it traded just four days – all up then exited the market (won’t buy back until 9/30) .   But as Bob points out they can also stop working in any given year.  We had a bad July, for example, as the bottom was later than projected.  Still the strategy continues to outperform its benchmark – the DJIA - for the year.  Very simple concept but it appears to continue to be effective.  Jerry

                        psi

                         

                        The Political Seasonality Index(PSI) is shown for information sake, as the author published it for a number of years in Barron's.  It represents the average of 13 political and seasonal tendencies of the market.  Drawn on over 100 years of Dow Jones Industrial Average data it projects one year in advance.  It is useful not for its trending direction (the market has on average gone up over the last century so the PSI always trends up for any 12 month period) but rather for the dates of possible turning points.   Accounts can be traded based on the PSI peaks and valleys on most investment platforms including Strategic Solutions. Past Performance does not guarantee future investment success.  Inherent in any investment is the potential for loss as well as the potential for gain. A list of all recommendations made within the immediately preceding year is available upon written request.

                         

                        www.flexibleplan.com

                         

                        From: Behavioral-Finance@yahoogroups.com [mailto:Behavioral-Finance@yahoogroups.com] On Behalf Of Bob Bronson
                        Sent: Wednesday, September 08, 2010 12:54 AM
                        To: Behavioral-Finance@yahoogroups.com
                        Subject: RE: [Behavioral-Finance] Dow Monthly returns

                         




                        Sergey,

                         

                        I agree there are not too many degrees of (statistical) freedom in ferreting out

                        the monthly pattern of the Annual Cycle.  My point there was in reference to

                        using five factors, or conditional filters, to make my point about this September

                        most likely being a down month in spite of its first three day’s strong start. 

                         

                        Furthermore, it would not be surprising if Friday’s stock market intraday high

                        was/is the high for the month, especially if tomorrow’s intraday action doesn’t

                        make a higher high.  But if does, it probably will be only a slightly higher high

                        and then it will probably net decline over the rest of the month.  By the way,

                        on a daily rather than only month-end pricing basis, October 7 is the historical

                        average low point, especially at the end of the Four Year Cycle.

                         

                        As far as the stock market changing its structure over time, consider that some

                        persistent anomalies, or non-random patterns, like the monthly pattern of the

                        annual cycle, may be robust, but aperiodically vary because of market efficiency,

                        or what we call competitive participant-observer feedback (gaming) on such

                        everybody-knows patterns.  We have noted before that the six-to-seven month

                        Strong Seasons and five-to-six month Weak Seasons of the Annual Cycle (not

                        calendar) are very robust because they are fundamentally grounded and are

                        not likely to disappear, at least due to market efficiency, but they certainly

                        exhibit pattern alternation (recall the Oct 2007 high and the Mar 2009 low).

                         

                        We’ve also pointed out to those on our private email list that some shorter term

                        calendar patterns, or time cycles, like the Yearend Holiday Season patterns,

                        which has been socialized into existence, are so gamed that they pattern

                        alternate being bigger or smaller than their historical average.  In this way

                        they still can maintain their historical average on an ex post basis, but vary

                        from it sufficiently to be non-profitable for those who simply play the average

                        pattern.  While, I believe this partially explains your observation of the varying

                        intensity of the Annual Cycle illustrated in your chart below, but I, for one,

                        would like to know more about your crisis theory of the variation in their activity.

                         

                        Conceptually and empirically, even institutionalized (e.g., quarterly financial

                        reporting) and socialized (e.g. holiday) time cycles can become somewhat of

                        a non-occurrence because of such feedback-driven pattern alteration, in which

                        case we say they exhibit “meaningless means.”  This apparent oddity is kind

                        of like the center of gravity being an out-of-body point like with a hula hoop

                        or a cantilever.  Similarly, and despite their being common in mathematics,

                        asymptotic patterns are extremely rare in the capital markets, so we prefer

                        to call the most powerful pricing force reversion-to-the-extreme behavior.

                         

                        In other words, the market’s efficiency does not require that a non-random

                        pattern’s average performance completely disappear, but rather it will become

                        predictably non-profitable, even if it is fundamentally grounded like the Strong

                        and Weak Seasons of the Annual Cycle, the Yearend Holiday Season and

                        the Four Year Cycle (recall the year late occurrence in 1986-7 and the cycle

                        inversion in 2006).

                         

                        Pattern alternation is one of several reasons why we believe the Semi-Weak

                        Case of the Efficient Market Hypothesis is the most operative, as I’ve pointed

                        out before in this forum when I noted there are consistent A players, like in

                        most any game as trading/investing surely is, despite random-walk arguments

                        to the contrary. 

                         

                        In this regard, I will be soon sending to our private email list (which anyone

                        here can join) a report, Moody Market, that presents heretofore unpublished

                        (as far as we know) aspects of the distribution of price returns over many time

                        frames that are very interesting.  They can be explained by the trend-following

                        (herding) behavior of institutional and individual investors and traders.

                         

                         

                         

                        _________________________________________________________________

                        From: Behavioral-Finance@yahoogroups.com [mailto:Behavioral-Finance@yahoogroups.com]

                        On Behalf Of SergeyTS
                        Sent: Tuesday, September 07, 2010 6:21 PM
                        To: Behavioral-Finance@yahoogroups.com
                        Subject: Re: [Behavioral-Finance] Dow Monthly returns

                         

                         

                        Hello, Bob:

                         

                        Thank you, this is a very interesting information.

                         

                        I want to add just a few words re Annual cycle. Actually it has not many degrees of freedom, and the overfitting there is not possible. This is a very simple model.

                         

                        But - I would not go there now. What I am more interested to discuss in regards to Technical Analysis is other things.My point is that the stock market is changing its structure with the time. If we look at any indicator/factor within some significant period of time, we see that. The Annual cycle is just one way to see that. Consider the market and the Annual cycle now and 100 years ago. The Annual cycle was stronger/more obvious 100 years ago than now. (And there are fundamentals involved: agricultural sector now is 2% while then it was 40-50%  of GDP, something like that, I do not remember exact numbers). Also, I have observed that the Annual cycle becomes stronger (or, in other words, its role becomes more obvious) at crisis times. Maybe during crisis we all tend to go back to old things that worked for us in the apst, I do not know.

                         

                        Please look at this diagram, it shows the level of activity of the Annual cycle within the latest 120 years. You can see there that during the crisises of 1970s, in 1987, 2001 the Annual cycle was more active (more red on the diagram):

                         

                         

                        Best regards,

                         

                        Sergey

                         

                         

                        ----- Original Message -----

                        Sent: Monday, September 06, 2010 7:23 PM

                        Subject: RE: [Behavioral-Finance] Dow Monthly returns

                         

                         

                        Whoops, the referenced charts are now inserted below.

                        From: Behavioral-Finance@yahoogroups.com [mailto:Behavioral-Finance@yahoogroups.com] On Behalf Of Bob Bronson
                        Sent: Monday, September 06, 2010 4:53 PM
                        To: Behavioral-Finance@yahoogroups.com
                        Subject: RE: [Behavioral-Finance] Dow Monthly returns


                        Thanx, Sergey. 

                        Here are the charts for the calendar-year monthly pattern using price-only, total return (dividends reinvested)

                        and real total return (dividends reinvested and adjusted for price inflation) using our CMI back to 1870.  Note

                        that the three charts illustrate the average (mean) in red and the statistically more significant median in blue.

                        The lower pair of red and blue lines in each chart is net effect of detrending the upper pair, which illustrates

                        the monthly pattern after the bullish effect of the annualized return for the upper pair is removed.  In contrast

                        to your detrended chart, which I discerned you presented from reading your very interesting website, that

                        not only do the detrended pairs show that January has typically been higher that April, but that is especially

                        so with the median indexes.

                        However, your and our work agree that September has created the lowest intra-year month end on average,

                        whether the stock market is measured with or without dividends and whether inflation is considered or not.

                        In further follow up to my previous comments below, I was using five factors in my analysis: (a) stock

                        market price; (b) (aggregated) dividends; (c) (price) inflation; (d) a simplistic form of price momentum,

                        or actually stock market price velocity - year-to-date net price change through August; and (e) our notion

                        of Supercycle Periods, which we quantify both fundamentally and technically, as was summarized there.

                        (As most everyone here knows, physicists define momentum as mass times velocity, a vector, while most

                        investors using price momentum are simply referring to a scalar metric, the rate of change of price, or its

                        velocity: the first derivative, or first differences of discrete a price time series).)

                        I say a simplistic form of momentum since as a scalar metric, I made my point about September’s history

                        having been a down month using Bayesian if-then conditional logic, which quantifies independent variables

                        (conditional probabilities) and dependent variables (unconditional probabilities) in a simplistic dichotomous

                        manner of simply true or false, which in this case is monthly returns being up (positive returns) or down

                        (negative returns).  The magnitude of those variables is entirely ignored. 

                        If I presented a scattergram, or regression analysis, of simply using the monthly returns data as vector

                        (with both plus or minus direction and the associated magnitude) a more quantitative case can be made

                        about September’s likelihood of having a negative return since both the odds would be greater and they

                        would be based on more than just the four instances I cited, wherein I introduced Supercycle Periods as

                        a qualifying factor.  But in spite of the higher degrees of freedom, which some might wrong-headedly

                        believe is (excess) data mining, of over fitting, we prefer the more qualitative approach as suggested

                        by the confluence of confirming factors from indicators in our four-paired factors forecasting models

                        that quantify monetary-economic, socio-political, valuation-sentiment and intra/inter market indicators.

                        Finally, I note that if interest rates are added to the five factor analysis I introduced below, the probability

                        for September having a negative return, and especially from now forward, would be even stronger,

                        even just using Bayesian analysis of the historical data since interest rates lead price inflation.  See

                        those lead-lag comparison charts near the bottom of this article:

                        Our August 10, 2010 Call: The Summer Stock Market Rally Is Over

                        http://www.financialsense.com/contributors/bob-bronson/the-summer-stock-market-rally-is-over

                        From: Behavioral-Finance@yahoogroups.com [mailto:Behavioral-Finance@yahoogroups.com] On Behalf Of SergeyTS
                        Sent: Monday, September 06, 2010 7:49 AM
                        To: Behavioral-Finance@yahoogroups.com
                        Subject: Re: [Behavioral-Finance] Dow Monthly returns

                        Hello,

                        This is Annual cycle for Dow:

                        the price history since 1789 year is used. Actually, Dow exists since 1885, before that the data are the interpolated prices of the most  liquid companies.

                        Best regards.

                        Sergey.

                        ----- Original Message -----

                        Sent: Monday, September 06, 2010 5:04 AM

                        Subject: RE: [Behavioral-Finance] Dow Monthly returns

                         

                        Based upon the Dow’s price-only history as Mark has referenced below, when the year-to-date

                        through August was down, as was the situation this year, September has been down about twice

                        as often as it has been up.

                        However, using total return (dividends reinvested) with our monthly CMI database for the 139

                        years since 1870 (S&P 90 and 500 indexes starting in Jan ’26 and Jan ’50, respectively, and

                        S&P-Wilshire 5000 since Apr ’01) – available upon request – there have been 45 instances with

                        their years-to-date down through August, of which 28, or 62%, Septembers had negative returns.

                        Furthermore, during the 12 second Presidential-term years since 1962, when the 48-month,

                        socio-political-economic, or Kitchin, cycle fully emerged (see footnotes # 4, 10, 14 and 20

                        here:  A Forecasting Model That Integrates Multiple Business and Stock Market Cycles

                        September has been down 67% of the time, or twice as often as it has been up: eight times

                        down compared to only four times up. 

                        Considering only Septembers when both it was a Presidential term second year, typically the

                        weakest of the four years (the political discord and economic uncertainty this year is unusually

                        bearish) and the stock market, measured by our CMI, was down year-to-date through August,

                        five of those seven instances were down or 71% of the time.

                        Finally, four of those seven were during Supercycle Bear Market Periods (e.g., 1881-1896,

                        1906-21, 1929-49, 1966-82 and 2000 to date) of which three, or 75%, of those Septembers

                        had negative returns - 1966, 1974 and 2002 , while September 2006 gained 2.3%.  

                        Since our work clearly shows that we’re still in a Supercycle Bear Market Period, in particular,

                        the most devastating of the two kinds: a deflationary economic Supercycle Winter – see the

                        summary of our quantified Supercycle Economic Seasons below - we believe the odds that

                        this September, while it has started off with a strong start, will decline for the rest of the month,

                        and it will probably even finish with a net negative return. 

                        Of course, we have other reasons for expecting this also – see the summary of our forecasting

                        models further below.

                        Bob Bronson

                        Bronson Capital Markets Research

                        http://financialsense.com/editorials/bronson/main.html

                        http://www.financialsense.com/contributors/bob-bronson

                        http://www.financialsense.com/sites/default/files/users/u188/images/2010/supercycle-economic-seasons.jpg

                        http://www.financialsense.com/sites/default/files/users/u188/images/2010/bronson-forecast-models.jpg

                        -----Original Message-----
                        From: Behavioral-Finance@yahoogroups.com [mailto:Behavioral-Finance@yahoogroups.com] On Behalf Of mark
                        Sent: Sunday, September 05, 2010 1:49 PM
                        To: Behavioral-Finance@yahoogroups.com
                        Subject: [Behavioral-Finance] Dow Monthly returns

                        This chart shows the Dow Industrials returns on a monthly basis back to 1886.  It is sorted on September returns

                        http://www.mutualfundsystem.com/newsletters/DowMonthlyReturns.pdf






                         




                      • Kwok Yeung
                        Dear Bob, There are a lot of interesting discussions on different cycles. But I think in this group, there are many members who are more interested in
                        Message 11 of 15 , Sep 8, 2010
                        View Source
                        • 0 Attachment
                          Dear Bob,

                          There are a lot of interesting discussions on different cycles.

                          But I think in this group, there are many members who are more interested in understanding/discussing how the BF factors or market forces that explain why some cycles work and some do not in different time. Similar to the case that Professor W. Jevons used sunspot activities to explain the Trade Cycle more than a century ago.  

                          Following the above line of thinking, it is nice to see that you mentioned at the end of the last email that:
                          " They can be explained by the trend-following(herding) behavior of institutional and individual investors and traders."

                          Do you think herding behavior can explain all the cases of why some cycles work and some do not? Also what factors would drive the crowd to reverse the direction?

                          Regards,

                          K.C. 






                          --- On Wed, 9/8/10, Bob Bronson <bob@...> wrote:

                          From: Bob Bronson <bob@...>
                          Subject: RE: [Behavioral-Finance] Dow Monthly returns
                          To: Behavioral-Finance@yahoogroups.com
                          Date: Wednesday, September 8, 2010, 4:54 AM

                           

                          Sergey,

                           

                          I agree there are not too many degrees of (statistical) freedom in ferreting out

                          the monthly pattern of the Annual Cycle.  My point there was in reference to

                          using five factors, or conditional filters, to make my point about this September

                          most likely being a down month in spite of its first three day’s strong start. 

                           

                          Furthermore, it would not be surprising if Friday’s stock market intraday high

                          was/is the high for the month, especially if tomorrow’s intraday action doesn’t

                          make a higher high.  But if does, it probably will be only a slightly higher high

                          and then it will probably net decline over the rest of the month.  By the way,

                          on a daily rather than only month-end pricing basis, October 7 is the historical

                          average low point, especially at the end of the Four Year Cycle.

                           

                          As far as the stock market changing its structure over time, consider that some

                          persistent anomalies, or non-random patterns, like the monthly pattern of the

                          annual cycle, may be robust, but aperiodically vary because of market efficiency,

                          or what we call competitive participant-observer feedback (gaming) on such

                          everybody-knows patterns.  We have noted before that the six-to-seven month

                          Strong Seasons and five-to-six month Weak Seasons of the Annual Cycle (not

                          calendar) are very robust because they are fundamentally grounded and are

                          not likely to disappear, at least due to market efficiency, but they certainly

                          exhibit pattern alternation (recall the Oct 2007 high and the Mar 2009 low).

                           

                          We’ve also pointed out to those on our private email list that some shorter term

                          calendar patterns, or time cycles, like the Yearend Holiday Season patterns,

                          which has been socialized into existence, are so gamed that they pattern

                          alternate being bigger or smaller than their historical average.  In this way

                          they still can maintain their historical average on an ex post basis, but vary

                          from it sufficiently to be non-profitable for those who simply play the average

                          pattern.  While, I believe this partially explains your observation of the varying

                          intensity of the Annual Cycle illustrated in your chart below, but I, for one,

                          would like to know more about your crisis theory of the variation in their activity.

                           

                          Conceptually and empirically, even institutionalized (e.g., quarterly financial

                          reporting) and socialized (e.g. holiday) time cycles can become somewhat of

                          a non-occurrence because of such feedback-driven pattern alteration, in which

                          case we say they exhibit “meaningless means.”  This apparent oddity is kind

                          of like the center of gravity being an out-of-body point like with a hula hoop

                          or a cantilever.  Similarly, and despite their being common in mathematics,

                          asymptotic patterns are extremely rare in the capital markets, so we prefer

                          to call the most powerful pricing force reversion-to-the-extreme behavior.

                           

                          In other words, the market’s efficiency does not require that a non-random

                          pattern’s average performance completely disappear, but rather it will become

                          predictably non-profitable, even if it is fundamentally grounded like the Strong

                          and Weak Seasons of the Annual Cycle, the Yearend Holiday Season and

                          the Four Year Cycle (recall the year late occurrence in 1986-7 and the cycle

                          inversion in 2006).

                           

                          Pattern alternation is one of several reasons why we believe the Semi-Weak

                          Case of the Efficient Market Hypothesis is the most operative, as I’ve pointed

                          out before in this forum when I noted there are consistent A players, like in

                          most any game as trading/investing surely is, despite random-walk arguments

                          to the contrary. 

                           

                          In this regard, I will be soon sending to our private email list (which anyone

                          here can join) a report, Moody Market, that presents heretofore unpublished

                          (as far as we know) aspects of the distribution of price returns over many time

                          frames that are very interesting.  They can be explained by the trend-following

                          (herding) behavior of institutional and individual investors and traders.

                           

                           

                           

                          _________________________________________________________________

                          From: Behavioral-Finance@yahoogroups.com [mailto:Behavioral-Finance@yahoogroups.com]

                          On Behalf Of SergeyTS
                          Sent: Tuesday, September 07, 2010 6:21 PM
                          To: Behavioral-Finance@yahoogroups.com
                          Subject: Re: [Behavioral-Finance] Dow Monthly returns

                           

                           

                          Hello, Bob:

                           

                          Thank you, this is a very interesting information.

                           

                          I want to add just a few words re Annual cycle. Actually it has not many degrees of freedom, and the overfitting there is not possible. This is a very simple model.

                           

                          But - I would not go there now. What I am more interested to discuss in regards to Technical Analysis is other things.My point is that the stock market is changing its structure with the time. If we look at any indicator/factor within some significant period of time, we see that. The Annual cycle is just one way to see that. Consider the market and the Annual cycle now and 100 years ago. The Annual cycle was stronger/more obvious 100 years ago than now. (And there are fundamentals involved: agricultural sector now is 2% while then it was 40-50%  of GDP, something like that, I do not remember exact numbers). Also, I have observed that the Annual cycle becomes stronger (or, in other words, its role becomes more obvious) at crisis times. Maybe during crisis we all tend to go back to old things that worked for us in the apst, I do not know.

                           

                          Please look at this diagram, it shows the level of activity of the Annual cycle within the latest 120 years. You can see there that during the crisises of 1970s, in 1987, 2001 the Annual cycle was more active (more red on the diagram):

                           

                           

                          Best regards,

                           

                          Sergey

                           

                           

                          ----- Original Message -----

                          Sent: Monday, September 06, 2010 7:23 PM

                          Subject: RE: [Behavioral-Finance] Dow Monthly returns

                           

                           

                          Whoops, the referenced charts are now inserted below.

                          From: Behavioral-Finance@yahoogroups.com [mailto:Behavioral-Finance@yahoogroups.com] On Behalf Of Bob Bronson
                          Sent: Monday, September 06, 2010 4:53 PM
                          To: Behavioral-Finance@yahoogroups.com
                          Subject: RE: [Behavioral-Finance] Dow Monthly returns



                          Thanx, Sergey. 

                          Here are the charts for the calendar-year monthly pattern using price-only, total return (dividends reinvested)

                          and real total return (dividends reinvested and adjusted for price inflation) using our CMI back to 1870.  Note

                          that the three charts illustrate the average (mean) in red and the statistically more significant median in blue.

                          The lower pair of red and blue lines in each chart is net effect of detrending the upper pair, which illustrates

                          the monthly pattern after the bullish effect of the annualized return for the upper pair is removed.  In contrast

                          to your detrended chart, which I discerned you presented from reading your very interesting website, that

                          not only do the detrended pairs show that January has typically been higher that April, but that is especially

                          so with the median indexes.

                          However, your and our work agree that September has created the lowest intra-year month end on average,

                          whether the stock market is measured with or without dividends and whether inflation is considered or not.

                          In further follow up to my previous comments below, I was using five factors in my analysis: (a) stock

                          market price; (b) (aggregated) dividends; (c) (price) inflation; (d) a simplistic form of price momentum,

                          or actually stock market price velocity - year-to-date net price change through August; and (e) our notion

                          of Supercycle Periods, which we quantify both fundamentally and technically, as was summarized there.

                          (As most everyone here knows, physicists define momentum as mass times velocity, a vector, while most

                          investors using price momentum are simply referring to a scalar metric, the rate of change of price, or its

                          velocity: the first derivative, or first differences of discrete a price time series).)

                          I say a simplistic form of momentum since as a scalar metric, I made my point about September’s history

                          having been a down month using Bayesian if-then conditional logic, which quantifies independent variables

                          (conditional probabilities) and dependent variables (unconditional probabilities) in a simplistic dichotomous

                          manner of simply true or false, which in this case is monthly returns being up (positive returns) or down

                          (negative returns).  The magnitude of those variables is entirely ignored. 

                          If I presented a scattergram, or regression analysis, of simply using the monthly returns data as vector

                          (with both plus or minus direction and the associated magnitude) a more quantitative case can be made

                          about September’s likelihood of having a negative return since both the odds would be greater and they

                          would be based on more than just the four instances I cited, wherein I introduced Supercycle Periods as

                          a qualifying factor.  But in spite of the higher degrees of freedom, which some might wrong-headedly

                          believe is (excess) data mining, of over fitting, we prefer the more qualitative approach as suggested

                          by the confluence of confirming factors from indicators in our four-paired factors forecasting models

                          that quantify monetary-economic, socio-political, valuation-sentiment and intra/inter market indicators.

                          Finally, I note that if interest rates are added to the five factor analysis I introduced below, the probability

                          for September having a negative return, and especially from now forward, would be even stronger,

                          even just using Bayesian analysis of the historical data since interest rates lead price inflation.  See

                          those lead-lag comparison charts near the bottom of this article:

                          Our August 10, 2010 Call: The Summer Stock Market Rally Is Over

                          http://www.financialsense.com/contributors/bob-bronson/the-summer-stock-market-rally-is-over

                          From: Behavioral-Finance@yahoogroups.com [mailto:Behavioral-Finance@yahoogroups.com] On Behalf Of SergeyTS
                          Sent: Monday, September 06, 2010 7:49 AM
                          To: Behavioral-Finance@yahoogroups.com
                          Subject: Re: [Behavioral-Finance] Dow Monthly returns

                          Hello,

                          This is Annual cycle for Dow:

                          the price history since 1789 year is used. Actually, Dow exists since 1885, before that the data are the interpolated prices of the most  liquid companies.

                          Best regards.

                          Sergey.

                          ----- Original Message -----

                          Sent: Monday, September 06, 2010 5:04 AM

                          Subject: RE: [Behavioral-Finance] Dow Monthly returns

                           

                          Based upon the Dow’s price-only history as Mark has referenced below, when the year-to-date

                          through August was down, as was the situation this year, September has been down about twice

                          as often as it has been up.

                          However, using total return (dividends reinvested) with our monthly CMI database for the 139

                          years since 1870 (S&P 90 and 500 indexes starting in Jan ’26 and Jan ’50, respectively, and

                          S&P-Wilshire 5000 since Apr ’01) – available upon request – there have been 45 instances with

                          their years-to-date down through August, of which 28, or 62%, Septembers had negative returns.

                          Furthermore, during the 12 second Presidential-term years since 1962, when the 48-month,

                          socio-political-economic, or Kitchin, cycle fully emerged (see footnotes # 4, 10, 14 and 20

                          here:  A Forecasting Model That Integrates Multiple Business and Stock Market Cycles

                          September has been down 67% of the time, or twice as often as it has been up: eight times

                          down compared to only four times up. 

                          Considering only Septembers when both it was a Presidential term second year, typically the

                          weakest of the four years (the political discord and economic uncertainty this year is unusually

                          bearish) and the stock market, measured by our CMI, was down year-to-date through August,

                          five of those seven instances were down or 71% of the time.

                          Finally, four of those seven were during Supercycle Bear Market Periods (e.g., 1881-1896,

                          1906-21, 1929-49, 1966-82 and 2000 to date) of which three, or 75%, of those Septembers

                          had negative returns - 1966, 1974 and 2002 , while September 2006 gained 2.3%.  

                          Since our work clearly shows that we’re still in a Supercycle Bear Market Period, in particular,

                          the most devastating of the two kinds: a deflationary economic Supercycle Winter – see the

                          summary of our quantified Supercycle Economic Seasons below - we believe the odds that

                          this September, while it has started off with a strong start, will decline for the rest of the month,

                          and it will probably even finish with a net negative return. 

                          Of course, we have other reasons for expecting this also – see the summary of our forecasting

                          models further below.

                          Bob Bronson

                          Bronson Capital Markets Research

                          http://financialsense.com/editorials/bronson/main.html

                          http://www.financialsense.com/contributors/bob-bronson

                          http://www.financialsense.com/sites/default/files/users/u188/images/2010/supercycle-economic-seasons.jpg

                          http://www.financialsense.com/sites/default/files/users/u188/images/2010/bronson-forecast-models.jpg

                          -----Original Message-----
                          From: Behavioral-Finance@yahoogroups.com [mailto:Behavioral-Finance@yahoogroups.com] On Behalf Of mark
                          Sent: Sunday, September 05, 2010 1:49 PM
                          To: Behavioral-Finance@yahoogroups.com
                          Subject: [Behavioral-Finance] Dow Monthly returns

                          This chart shows the Dow Industrials returns on a monthly basis back to 1886.  It is sorted on September returns

                          http://www.mutualfundsystem.com/newsletters/DowMonthlyReturns.pdf








                        • Bob Bronson
                          Yes, K.C., trend-following (herding behavior)r mostly explains the fail tailed distributions of returns over all time horizons. Of course, the predominance of
                          Message 12 of 15 , Sep 8, 2010
                          View Source
                          • 0 Attachment

                            Yes, K.C., trend-following (herding behavior)r mostly explains the fail tailed distributions

                            of returns over all time horizons.  Of course, the predominance of risk aversion causes

                            the negative skewness of generally bigger negative than positive tails.  And under- and

                            over-estimation causes flatter tops (i.e., negative kurtosis) in most of those distributions.

                             

                            But aspects of the predictability of the cyclical anomalies that reflect these distribution

                            patterns is what leads to investment profits and thus is most interesting, don’t you agree?

                             

                            Bob

                             

                             

                            From: Behavioral-Finance@yahoogroups.com [mailto:Behavioral-Finance@yahoogroups.com] On Behalf Of Kwok Yeung
                            Sent: Wednesday, September 08, 2010 1:22 PM
                            To: Behavioral-Finance@yahoogroups.com
                            Subject: RE: [Behavioral-Finance] Dow Monthly returns

                             

                            Dear Bob,

                            There are a lot of interesting discussions on different cycles.

                            But I think in this group, there are many members who are more interested in understanding/discussing how the BF factors or market forces that explain why some cycles work and some do not in different time. Similar to the case that Professor W. Jevons used sunspot activities to explain the Trade Cycle more than a century ago.  

                            Following the above line of thinking, it is nice to see that you mentioned at the end of the last email that:
                            " They can be explained by the trend-following (herding) behavior of institutional and individual investors and traders."

                            That’s why I posted those comments, especially following Peter’s comments.

                            Do you think herding behavior can explain all the cases of why some cycles work and some do not? Also what factors would drive the crowd to reverse the direction?

                            Regards,

                            K.C. 


                            --- On Wed, 9/8/10, Bob Bronson <bob@...> wrote:


                            From: Bob Bronson <bob@...>
                            Subject: RE: [Behavioral-Finance] Dow Monthly returns
                            To: Behavioral-Finance@yahoogroups.com
                            Date: Wednesday, September 8, 2010, 4:54 AM

                             

                            Sergey,

                             

                            I agree there are not too many degrees of (statistical) freedom in ferreting out

                            the monthly pattern of the Annual Cycle.  My point there was in reference to

                            using five factors, or conditional filters, to make my point about this September

                            most likely being a down month in spite of its first three day’s strong start. 

                             

                            Furthermore, it would not be surprising if Friday’s stock market intraday high

                            was/is the high for the month, especially if tomorrow’s intraday action doesn’t

                            make a higher high.  But if does, it probably will be only a slightly higher high

                            and then it will probably net decline over the rest of the month.  By the way,

                            on a daily rather than only month-end pricing basis, October 7 is the historical

                            average low point, especially at the end of the Four Year Cycle.

                             

                            As far as the stock market changing its structure over time, consider that some

                            persistent anomalies, or non-random patterns, like the monthly pattern of the

                            annual cycle, may be robust, but aperiodically vary because of market efficiency,

                            or what we call competitive participant-observer feedback (gaming) on such

                            everybody-knows patterns.  We have noted before that the six-to-seven month

                            Strong Seasons and five-to-six month Weak Seasons of the Annual Cycle (not

                            calendar) are very robust because they are fundamentally grounded and are

                            not likely to disappear, at least due to market efficiency, but they certainly

                            exhibit pattern alternation (recall the Oct 2007 high and the Mar 2009 low).

                             

                            We’ve also pointed out to those on our private email list that some shorter term

                            calendar patterns, or time cycles, like the Yearend Holiday Season patterns,

                            which has been socialized into existence, are so gamed that they pattern

                            alternate being bigger or smaller than their historical average.  In this way

                            they still can maintain their historical average on an ex post basis, but vary

                            from it sufficiently to be non-profitable for those who simply play the average

                            pattern.  While, I believe this partially explains your observation of the varying

                            intensity of the Annual Cycle illustrated in your chart below, but I, for one,

                            would like to know more about your crisis theory of the variation in their activity.

                             

                            Conceptually and empirically, even institutionalized (e.g., quarterly financial

                            reporting) and socialized (e.g. holiday) time cycles can become somewhat of

                            a non-occurrence because of such feedback-driven pattern alteration, in which

                            case we say they exhibit “meaningless means.”  This apparent oddity is kind

                            of like the center of gravity being an out-of-body point like with a hula hoop

                            or a cantilever.  Similarly, and despite their being common in mathematics,

                            asymptotic patterns are extremely rare in the capital markets, so we prefer

                            to call the most powerful pricing force reversion-to-the-extreme behavior.

                             

                            In other words, the market’s efficiency does not require that a non-random

                            pattern’s average performance completely disappear, but rather it will become

                            predictably non-profitable, even if it is fundamentally grounded like the Strong

                            and Weak Seasons of the Annual Cycle, the Yearend Holiday Season and

                            the Four Year Cycle (recall the year late occurrence in 1986-7 and the cycle

                            inversion in 2006).

                             

                            Pattern alternation is one of several reasons why we believe the Semi-Weak

                            Case of the Efficient Market Hypothesis is the most operative, as I’ve pointed

                            out before in this forum when I noted there are consistent A players, like in

                            most any game as trading/investing surely is, despite random-walk arguments

                            to the contrary. 

                             

                            In this regard, I will be soon sending to our private email list (which anyone

                            here can join) a report, Moody Market, that presents heretofore unpublished

                            (as far as we know) aspects of the distribution of price returns over many time

                            frames that are very interesting.  They can be explained by the trend-following

                            (herding) behavior of institutional and individual investors and traders.

                             

                             

                             

                            _________________________________________________________________

                            From: Behavioral-Finance@yahoogroups.com [mailto:Behavioral-Finance@yahoogroups.com]

                            On Behalf Of SergeyTS
                            Sent: Tuesday, September 07, 2010 6:21 PM
                            To: Behavioral-Finance@yahoogroups.com
                            Subject: Re: [Behavioral-Finance] Dow Monthly returns

                             

                             

                            Hello, Bob:

                             

                            Thank you, this is a very interesting information.

                             

                            I want to add just a few words re Annual cycle. Actually it has not many degrees of freedom, and the overfitting there is not possible. This is a very simple model.

                             

                            But - I would not go there now. What I am more interested to discuss in regards to Technical Analysis is other things.My point is that the stock market is changing its structure with the time. If we look at any indicator/factor within some significant period of time, we see that. The Annual cycle is just one way to see that. Consider the market and the Annual cycle now and 100 years ago. The Annual cycle was stronger/more obvious 100 years ago than now. (And there are fundamentals involved: agricultural sector now is 2% while then it was 40-50%  of GDP, something like that, I do not remember exact numbers). Also, I have observed that the Annual cycle becomes stronger (or, in other words, its role becomes more obvious) at crisis times. Maybe during crisis we all tend to go back to old things that worked for us in the apst, I do not know.

                             

                            Please look at this diagram, it shows the level of activity of the Annual cycle within the latest 120 years. You can see there that during the crisises of 1970s, in 1987, 2001 the Annual cycle was more active (more red on the diagram):

                             

                             

                            Best regards,

                             

                            Sergey

                             

                             

                            ----- Original Message -----

                            Sent: Monday, September 06, 2010 7:23 PM

                            Subject: RE: [Behavioral-Finance] Dow Monthly returns

                             

                             

                            Whoops, the referenced charts are now inserted below.

                            From: Behavioral-Finance@yahoogroups.com [mailto:Behavioral-Finance@yahoogroups.com] On Behalf Of Bob Bronson
                            Sent: Monday, September 06, 2010 4:53 PM
                            To: Behavioral-Finance@yahoogroups.com
                            Subject: RE: [Behavioral-Finance] Dow Monthly returns


                            Thanx, Sergey. 

                            Here are the charts for the calendar-year monthly pattern using price-only, total return (dividends reinvested)

                            and real total return (dividends reinvested and adjusted for price inflation) using our CMI back to 1870.  Note

                            that the three charts illustrate the average (mean) in red and the statistically more significant median in blue.

                            The lower pair of red and blue lines in each chart is net effect of detrending the upper pair, which illustrates

                            the monthly pattern after the bullish effect of the annualized return for the upper pair is removed.  In contrast

                            to your detrended chart, which I discerned you presented from reading your very interesting website, that

                            not only do the detrended pairs show that January has typically been higher that April, but that is especially

                            so with the median indexes.

                            However, your and our work agree that September has created the lowest intra-year month end on average,

                            whether the stock market is measured with or without dividends and whether inflation is considered or not.

                            In further follow up to my previous comments below, I was using five factors in my analysis: (a) stock

                            market price; (b) (aggregated) dividends; (c) (price) inflation; (d) a simplistic form of price momentum,

                            or actually stock market price velocity - year-to-date net price change through August; and (e) our notion

                            of Supercycle Periods, which we quantify both fundamentally and technically, as was summarized there.

                            (As most everyone here knows, physicists define momentum as mass times velocity, a vector, while most

                            investors using price momentum are simply referring to a scalar metric, the rate of change of price, or its

                            velocity: the first derivative, or first differences of discrete a price time series).)

                            I say a simplistic form of momentum since as a scalar metric, I made my point about September’s history

                            having been a down month using Bayesian if-then conditional logic, which quantifies independent variables

                            (conditional probabilities) and dependent variables (unconditional probabilities) in a simplistic dichotomous

                            manner of simply true or false, which in this case is monthly returns being up (positive returns) or down

                            (negative returns).  The magnitude of those variables is entirely ignored. 

                            If I presented a scattergram, or regression analysis, of simply using the monthly returns data as vector

                            (with both plus or minus direction and the associated magnitude) a more quantitative case can be made

                            about September’s likelihood of having a negative return since both the odds would be greater and they

                            would be based on more than just the four instances I cited, wherein I introduced Supercycle Periods as

                            a qualifying factor.  But in spite of the higher degrees of freedom, which some might wrong-headedly

                            believe is (excess) data mining, of over fitting, we prefer the more qualitative approach as suggested

                            by the confluence of confirming factors from indicators in our four-paired factors forecasting models

                            that quantify monetary-economic, socio-political, valuation-sentiment and intra/inter market indicators.

                            Finally, I note that if interest rates are added to the five factor analysis I introduced below, the probability

                            for September having a negative return, and especially from now forward, would be even stronger,

                            even just using Bayesian analysis of the historical data since interest rates lead price inflation.  See

                            those lead-lag comparison charts near the bottom of this article:

                            Our August 10, 2010 Call: The Summer Stock Market Rally Is Over

                            http://www.financialsense.com/contributors/bob-bronson/the-summer-stock-market-rally-is-over

                            From: Behavioral-Finance@yahoogroups.com [mailto:Behavioral-Finance@yahoogroups.com] On Behalf Of SergeyTS
                            Sent: Monday, September 06, 2010 7:49 AM
                            To: Behavioral-Finance@yahoogroups.com
                            Subject: Re: [Behavioral-Finance] Dow Monthly returns

                            Hello,

                            This is Annual cycle for Dow:

                            the price history since 1789 year is used. Actually, Dow exists since 1885, before that the data are the interpolated prices of the most  liquid companies.

                            Best regards.

                            Sergey.

                            ----- Original Message -----

                            Sent: Monday, September 06, 2010 5:04 AM

                            Subject: RE: [Behavioral-Finance] Dow Monthly returns

                             

                            Based upon the Dow’s price-only history as Mark has referenced below, when the year-to-date

                            through August was down, as was the situation this year, September has been down about twice

                            as often as it has been up.

                            However, using total return (dividends reinvested) with our monthly CMI database for the 139

                            years since 1870 (S&P 90 and 500 indexes starting in Jan ’26 and Jan ’50, respectively, and

                            S&P-Wilshire 5000 since Apr ’01) – available upon request – there have been 45 instances with

                            their years-to-date down through August, of which 28, or 62%, Septembers had negative returns.

                            Furthermore, during the 12 second Presidential-term years since 1962, when the 48-month,

                            socio-political-economic, or Kitchin, cycle fully emerged (see footnotes # 4, 10, 14 and 20

                            here:  A Forecasting Model That Integrates Multiple Business and Stock Market Cycles

                            September has been down 67% of the time, or twice as often as it has been up: eight times

                            down compared to only four times up. 

                            Considering only Septembers when both it was a Presidential term second year, typically the

                            weakest of the four years (the political discord and economic uncertainty this year is unusually

                            bearish) and the stock market, measured by our CMI, was down year-to-date through August,

                            five of those seven instances were down or 71% of the time.

                            Finally, four of those seven were during Supercycle Bear Market Periods (e.g., 1881-1896,

                            1906-21, 1929-49, 1966-82 and 2000 to date) of which three, or 75%, of those Septembers

                            had negative returns - 1966, 1974 and 2002 , while September 2006 gained 2.3%.  

                            Since our work clearly shows that we’re still in a Supercycle Bear Market Period, in particular,

                            the most devastating of the two kinds: a deflationary economic Supercycle Winter – see the

                            summary of our quantified Supercycle Economic Seasons below - we believe the odds that

                            this September, while it has started off with a strong start, will decline for the rest of the month,

                            and it will probably even finish with a net negative return. 

                            Of course, we have other reasons for expecting this also – see the summary of our forecasting

                            models further below.

                            Bob Bronson

                            Bronson Capital Markets Research

                            http://financialsense.com/editorials/bronson/main.html

                            http://www.financialsense.com/contributors/bob-bronson

                            http://www.financialsense.com/sites/default/files/users/u188/images/2010/supercycle-economic-seasons.jpg

                            http://www.financialsense.com/sites/default/files/users/u188/images/2010/bronson-forecast-models.jpg

                            -----Original Message-----
                            From: Behavioral-Finance@yahoogroups.com [mailto:Behavioral-Finance@yahoogroups.com] On Behalf Of mark
                            Sent: Sunday, September 05, 2010 1:49 PM
                            To: Behavioral-Finance@yahoogroups.com
                            Subject: [Behavioral-Finance] Dow Monthly returns

                            This chart shows the Dow Industrials returns on a monthly basis back to 1886.  It is sorted on September returns

                            http://www.mutualfundsystem.com/newsletters/DowMonthlyReturns.pdf






                             

                          • Jerry Wagner
                            Graph s attached From: Jerry Wagner [mailto:flexplan@ix.netcom.com] Sent: Wednesday, September 08, 2010 10:09 AM To: Behavioral-Finance@yahoogroups.com Cc:
                            Message 13 of 15 , Sep 8, 2010
                            View Source
                            • 1 Attachment
                            • 103 KB

                            Graph’s attached

                             

                            From: Jerry Wagner [mailto:flexplan@...]
                            Sent: Wednesday, September 08, 2010 10:09 AM
                            To: 'Behavioral-Finance@yahoogroups.com'
                            Cc: 'jerry@...'
                            Subject: RE: [Behavioral-Finance] Dow Monthly returns

                             

                            Bob and Sergey

                             

                            Below is a chart (the blue line) we publish in January of each year – I’ve been doing so for about 20 years.  Each week we chart the actual performance of the Dow (the red line) versus the projection.  We also have a strategy that trades its buy and sell points (which are available a year in advance).  Uncanny how accurate it can be at times (especially spring and fall).  Last week, for example, it traded just four days – all up then exited the market (won’t buy back until 9/30) .   But as Bob points out they can also stop working in any given year.  We had a bad July, for example, as the bottom was later than projected.  Still the strategy continues to outperform its benchmark – the DJIA - for the year.  Very simple concept but it appears to continue to be effective.  Jerry

                            psi

                             

                            The Political Seasonality Index(PSI) is shown for information sake, as the author published it for a number of years in Barron's.  It represents the average of 13 political and seasonal tendencies of the market.  Drawn on over 100 years of Dow Jones Industrial Average data it projects one year in advance.  It is useful not for its trending direction (the market has on average gone up over the last century so the PSI always trends up for any 12 month period) but rather for the dates of possible turning points.   Accounts can be traded based on the PSI peaks and valleys on most investment platforms including Strategic Solutions. Past Performance does not guarantee future investment success.  Inherent in any investment is the potential for loss as well as the potential for gain. A list of all recommendations made within the immediately preceding year is available upon written request.

                             

                            www.flexibleplan.com

                             

                            From: Behavioral-Finance@yahoogroups.com [mailto:Behavioral-Finance@yahoogroups.com] On Behalf Of Bob Bronson
                            Sent: Wednesday, September 08, 2010 12:54 AM
                            To: Behavioral-Finance@yahoogroups.com
                            Subject: RE: [Behavioral-Finance] Dow Monthly returns

                             



                            Sergey,

                             

                            I agree there are not too many degrees of (statistical) freedom in ferreting out

                            the monthly pattern of the Annual Cycle.  My point there was in reference to

                            using five factors, or conditional filters, to make my point about this September

                            most likely being a down month in spite of its first three day’s strong start. 

                             

                            Furthermore, it would not be surprising if Friday’s stock market intraday high

                            was/is the high for the month, especially if tomorrow’s intraday action doesn’t

                            make a higher high.  But if does, it probably will be only a slightly higher high

                            and then it will probably net decline over the rest of the month.  By the way,

                            on a daily rather than only month-end pricing basis, October 7 is the historical

                            average low point, especially at the end of the Four Year Cycle.

                             

                            As far as the stock market changing its structure over time, consider that some

                            persistent anomalies, or non-random patterns, like the monthly pattern of the

                            annual cycle, may be robust, but aperiodically vary because of market efficiency,

                            or what we call competitive participant-observer feedback (gaming) on such

                            everybody-knows patterns.  We have noted before that the six-to-seven month

                            Strong Seasons and five-to-six month Weak Seasons of the Annual Cycle (not

                            calendar) are very robust because they are fundamentally grounded and are

                            not likely to disappear, at least due to market efficiency, but they certainly

                            exhibit pattern alternation (recall the Oct 2007 high and the Mar 2009 low).

                             

                            We’ve also pointed out to those on our private email list that some shorter term

                            calendar patterns, or time cycles, like the Yearend Holiday Season patterns,

                            which has been socialized into existence, are so gamed that they pattern

                            alternate being bigger or smaller than their historical average.  In this way

                            they still can maintain their historical average on an ex post basis, but vary

                            from it sufficiently to be non-profitable for those who simply play the average

                            pattern.  While, I believe this partially explains your observation of the varying

                            intensity of the Annual Cycle illustrated in your chart below, but I, for one,

                            would like to know more about your crisis theory of the variation in their activity.

                             

                            Conceptually and empirically, even institutionalized (e.g., quarterly financial

                            reporting) and socialized (e.g. holiday) time cycles can become somewhat of

                            a non-occurrence because of such feedback-driven pattern alteration, in which

                            case we say they exhibit “meaningless means.”  This apparent oddity is kind

                            of like the center of gravity being an out-of-body point like with a hula hoop

                            or a cantilever.  Similarly, and despite their being common in mathematics,

                            asymptotic patterns are extremely rare in the capital markets, so we prefer

                            to call the most powerful pricing force reversion-to-the-extreme behavior.

                             

                            In other words, the market’s efficiency does not require that a non-random

                            pattern’s average performance completely disappear, but rather it will become

                            predictably non-profitable, even if it is fundamentally grounded like the Strong

                            and Weak Seasons of the Annual Cycle, the Yearend Holiday Season and

                            the Four Year Cycle (recall the year late occurrence in 1986-7 and the cycle

                            inversion in 2006).

                             

                            Pattern alternation is one of several reasons why we believe the Semi-Weak

                            Case of the Efficient Market Hypothesis is the most operative, as I’ve pointed

                            out before in this forum when I noted there are consistent A players, like in

                            most any game as trading/investing surely is, despite random-walk arguments

                            to the contrary. 

                             

                            In this regard, I will be soon sending to our private email list (which anyone

                            here can join) a report, Moody Market, that presents heretofore unpublished

                            (as far as we know) aspects of the distribution of price returns over many time

                            frames that are very interesting.  They can be explained by the trend-following

                            (herding) behavior of institutional and individual investors and traders.

                             

                             

                             

                            _________________________________________________________________

                            From: Behavioral-Finance@yahoogroups.com [mailto:Behavioral-Finance@yahoogroups.com]

                            On Behalf Of SergeyTS
                            Sent: Tuesday, September 07, 2010 6:21 PM
                            To: Behavioral-Finance@yahoogroups.com
                            Subject: Re: [Behavioral-Finance] Dow Monthly returns

                             

                             

                            Hello, Bob:

                             

                            Thank you, this is a very interesting information.

                             

                            I want to add just a few words re Annual cycle. Actually it has not many degrees of freedom, and the overfitting there is not possible. This is a very simple model.

                             

                            But - I would not go there now. What I am more interested to discuss in regards to Technical Analysis is other things.My point is that the stock market is changing its structure with the time. If we look at any indicator/factor within some significant period of time, we see that. The Annual cycle is just one way to see that. Consider the market and the Annual cycle now and 100 years ago. The Annual cycle was stronger/more obvious 100 years ago than now. (And there are fundamentals involved: agricultural sector now is 2% while then it was 40-50%  of GDP, something like that, I do not remember exact numbers). Also, I have observed that the Annual cycle becomes stronger (or, in other words, its role becomes more obvious) at crisis times. Maybe during crisis we all tend to go back to old things that worked for us in the apst, I do not know.

                             

                            Please look at this diagram, it shows the level of activity of the Annual cycle within the latest 120 years. You can see there that during the crisises of 1970s, in 1987, 2001 the Annual cycle was more active (more red on the diagram):

                             

                             

                            Best regards,

                             

                            Sergey

                             

                             

                            ----- Original Message -----

                            Sent: Monday, September 06, 2010 7:23 PM

                            Subject: RE: [Behavioral-Finance] Dow Monthly returns

                             

                             

                            Whoops, the referenced charts are now inserted below.

                            From: Behavioral-Finance@yahoogroups.com [mailto:Behavioral-Finance@yahoogroups.com] On Behalf Of Bob Bronson
                            Sent: Monday, September 06, 2010 4:53 PM
                            To: Behavioral-Finance@yahoogroups.com
                            Subject: RE: [Behavioral-Finance] Dow Monthly returns

                            Thanx, Sergey. 

                            Here are the charts for the calendar-year monthly pattern using price-only, total return (dividends reinvested)

                            and real total return (dividends reinvested and adjusted for price inflation) using our CMI back to 1870.  Note

                            that the three charts illustrate the average (mean) in red and the statistically more significant median in blue.

                            The lower pair of red and blue lines in each chart is net effect of detrending the upper pair, which illustrates

                            the monthly pattern after the bullish effect of the annualized return for the upper pair is removed.  In contrast

                            to your detrended chart, which I discerned you presented from reading your very interesting website, that

                            not only do the detrended pairs show that January has typically been higher that April, but that is especially

                            so with the median indexes.

                            However, your and our work agree that September has created the lowest intra-year month end on average,

                            whether the stock market is measured with or without dividends and whether inflation is considered or not.

                            In further follow up to my previous comments below, I was using five factors in my analysis: (a) stock

                            market price; (b) (aggregated) dividends; (c) (price) inflation; (d) a simplistic form of price momentum,

                            or actually stock market price velocity - year-to-date net price change through August; and (e) our notion

                            of Supercycle Periods, which we quantify both fundamentally and technically, as was summarized there.

                            (As most everyone here knows, physicists define momentum as mass times velocity, a vector, while most

                            investors using price momentum are simply referring to a scalar metric, the rate of change of price, or its

                            velocity: the first derivative, or first differences of discrete a price time series).)

                            I say a simplistic form of momentum since as a scalar metric, I made my point about September’s history

                            having been a down month using Bayesian if-then conditional logic, which quantifies independent variables

                            (conditional probabilities) and dependent variables (unconditional probabilities) in a simplistic dichotomous

                            manner of simply true or false, which in this case is monthly returns being up (positive returns) or down

                            (negative returns).  The magnitude of those variables is entirely ignored. 

                            If I presented a scattergram, or regression analysis, of simply using the monthly returns data as vector

                            (with both plus or minus direction and the associated magnitude) a more quantitative case can be made

                            about September’s likelihood of having a negative return since both the odds would be greater and they

                            would be based on more than just the four instances I cited, wherein I introduced Supercycle Periods as

                            a qualifying factor.  But in spite of the higher degrees of freedom, which some might wrong-headedly

                            believe is (excess) data mining, of over fitting, we prefer the more qualitative approach as suggested

                            by the confluence of confirming factors from indicators in our four-paired factors forecasting models

                            that quantify monetary-economic, socio-political, valuation-sentiment and intra/inter market indicators.

                            Finally, I note that if interest rates are added to the five factor analysis I introduced below, the probability

                            for September having a negative return, and especially from now forward, would be even stronger,

                            even just using Bayesian analysis of the historical data since interest rates lead price inflation.  See

                            those lead-lag comparison charts near the bottom of this article:

                            Our August 10, 2010 Call: The Summer Stock Market Rally Is Over

                            http://www.financialsense.com/contributors/bob-bronson/the-summer-stock-market-rally-is-over

                            From: Behavioral-Finance@yahoogroups.com [mailto:Behavioral-Finance@yahoogroups.com] On Behalf Of SergeyTS
                            Sent: Monday, September 06, 2010 7:49 AM
                            To: Behavioral-Finance@yahoogroups.com
                            Subject: Re: [Behavioral-Finance] Dow Monthly returns

                            Hello,

                            This is Annual cycle for Dow:

                            the price history since 1789 year is used. Actually, Dow exists since 1885, before that the data are the interpolated prices of the most  liquid companies.

                            Best regards.

                            Sergey.

                            ----- Original Message -----

                            Sent: Monday, September 06, 2010 5:04 AM

                            Subject: RE: [Behavioral-Finance] Dow Monthly returns

                             

                            Based upon the Dow’s price-only history as Mark has referenced below, when the year-to-date

                            through August was down, as was the situation this year, September has been down about twice

                            as often as it has been up.

                            However, using total return (dividends reinvested) with our monthly CMI database for the 139

                            years since 1870 (S&P 90 and 500 indexes starting in Jan ’26 and Jan ’50, respectively, and

                            S&P-Wilshire 5000 since Apr ’01) – available upon request – there have been 45 instances with

                            their years-to-date down through August, of which 28, or 62%, Septembers had negative returns.

                            Furthermore, during the 12 second Presidential-term years since 1962, when the 48-month,

                            socio-political-economic, or Kitchin, cycle fully emerged (see footnotes # 4, 10, 14 and 20

                            here:  A Forecasting Model That Integrates Multiple Business and Stock Market Cycles

                            September has been down 67% of the time, or twice as often as it has been up: eight times

                            down compared to only four times up. 

                            Considering only Septembers when both it was a Presidential term second year, typically the

                            weakest of the four years (the political discord and economic uncertainty this year is unusually

                            bearish) and the stock market, measured by our CMI, was down year-to-date through August,

                            five of those seven instances were down or 71% of the time.

                            Finally, four of those seven were during Supercycle Bear Market Periods (e.g., 1881-1896,

                            1906-21, 1929-49, 1966-82 and 2000 to date) of which three, or 75%, of those Septembers

                            had negative returns - 1966, 1974 and 2002 , while September 2006 gained 2.3%.  

                            Since our work clearly shows that we’re still in a Supercycle Bear Market Period, in particular,

                            the most devastating of the two kinds: a deflationary economic Supercycle Winter – see the

                            summary of our quantified Supercycle Economic Seasons below - we believe the odds that

                            this September, while it has started off with a strong start, will decline for the rest of the month,

                            and it will probably even finish with a net negative return. 

                            Of course, we have other reasons for expecting this also – see the summary of our forecasting

                            models further below.

                            Bob Bronson

                            Bronson Capital Markets Research

                            http://financialsense.com/editorials/bronson/main.html

                            http://www.financialsense.com/contributors/bob-bronson

                            http://www.financialsense.com/sites/default/files/users/u188/images/2010/supercycle-economic-seasons.jpg

                            http://www.financialsense.com/sites/default/files/users/u188/images/2010/bronson-forecast-models.jpg

                            -----Original Message-----
                            From: Behavioral-Finance@yahoogroups.com [mailto:Behavioral-Finance@yahoogroups.com] On Behalf Of mark
                            Sent: Sunday, September 05, 2010 1:49 PM
                            To: Behavioral-Finance@yahoogroups.com
                            Subject: [Behavioral-Finance] Dow Monthly returns

                            This chart shows the Dow Industrials returns on a monthly basis back to 1886.  It is sorted on September returns

                            http://www.mutualfundsystem.com/newsletters/DowMonthlyReturns.pdf



                             

                             



                          Your message has been successfully submitted and would be delivered to recipients shortly.