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

AI-GEOSTATS: Skipping nodes in GSLIB and 2 answers to a question

Expand Messages
  • geogis 123
    Dear all, many thanks to Pierre Goovaerts and Yetta Jager for answering my previous question (see below) I have another question (possibly pretty basic
    Message 1 of 3 , Jun 24, 2003
    • 0 Attachment
      Dear all,

      many thanks to Pierre Goovaerts and Yetta Jager for answering my previous
      question (see below)

      I have another question (possibly pretty basic ...apologies in case it is a
      RTFM question, but the manual is rather large :-)..

      I notice that irregular interpolated grids are published around (e.g. in
      P.Goovaerts book).

      Presumably nodes which are not of interest can be skipped by the relevant
      kriging program.
      Is there such an an option in GSLIB simulation programs, without altering
      the code? (i.e what needs to be set to what?)
      Alternatively, is there such an option in other software doing simulations?

      regards & thank you
      Frank




      <><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><>





      > Dear all,
      >
      > can someone help me with a a copule of related questions?
      > A possible approach to incorporate secondary data for estimation is to
      fit a
      > "traditional" statistical model using the variable of interest as the
      > dependent one, to estimate the trend, and to add to it interpolatated
      > residuals.
      >
      > 1)
      > Is this what Kriging with a variable mean is in GSLIB ?
      >
      > if this is the case, does it mean that the variogram to be supplied to
      > kriging and simulation programs is that of the residuals?
      >
      > 2)
      > Would it be sensible to adopt the approach described above using
      simulation
      > (rather than kriging)of residuals? My concern is that one might risk to
      come
      > up with realisations in which some locations have negative (and in my
      case
      > non sensical) estimates of the variable in question (i.e. how would one
      deal
      > with this)?
      >
      > thanks very much for your help
      >
      > Frank


      Hi,

      The answer is yes to your 1st question and I have never
      experienced problems with simulation using local mean.
      If it's performed in the normal space, the backtransform
      would allow one to restrict tthe range of simulated values
      to positive values.

      Regards,

      Pierre Goovaerts
      ---------------
      Dear Frank,

      Regarding your question to GEOSTATS, see my webpage publications for the
      report on PATTERN-PLUS, which basically combined GLS regression with
      kriged residuals. I guess the question of simulation vs. kriging depends
      on whether you want one most-likely answer at each location or a full
      distribution of possible answers.

      Yetta

      _________________________________________________________________
      It's fast, it's easy and it's free. Get MSN Messenger today!
      http://www.msn.co.uk/messenger


      --
      * To post a message to the list, send it to ai-geostats@...
      * As a general service to the users, please remember to post a summary of any useful responses to your questions.
      * To unsubscribe, send an email to majordomo@... with no subject and "unsubscribe ai-geostats" followed by "end" on the next line in the message body. DO NOT SEND Subscribe/Unsubscribe requests to the list
      * Support to the list is provided at http://www.ai-geostats.org
    • Pierre Goovaerts
      Hi Frank, You have to modify Gslib program. The way I do it is to read a grid of indicator values (1 for nodes you want to simulate and zero for the one to be
      Message 2 of 3 , Jun 24, 2003
      • 0 Attachment
        Hi Frank,

        You have to modify Gslib program. The way I do it is
        to read a grid of indicator values (1 for nodes you want to simulate
        and zero for the one to be skipped) and have a "if.. "
        command in Gslib to skip in the random path the selected nodes with
        a zero indicator value.

        Cheers,

        Pierre

        <><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><>

        Dr. Pierre Goovaerts
        President of PGeostat, LLC
        Chief Scientist with Biomedware Inc.
        710 Ridgemont Lane
        Ann Arbor, Michigan, 48103-1535, U.S.A.

        E-mail: goovaert@...
        Phone: (734) 668-9900
        Fax: (734) 668-7788
        http://alumni.engin.umich.edu/~goovaert/

        <><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><>

        On Tue, 24 Jun 2003, geogis 123 wrote:

        > Dear all,
        >
        > many thanks to Pierre Goovaerts and Yetta Jager for answering my previous
        > question (see below)
        >
        > I have another question (possibly pretty basic ...apologies in case it is a
        > RTFM question, but the manual is rather large :-)..
        >
        > I notice that irregular interpolated grids are published around (e.g. in
        > P.Goovaerts book).
        >
        > Presumably nodes which are not of interest can be skipped by the relevant
        > kriging program.
        > Is there such an an option in GSLIB simulation programs, without altering
        > the code? (i.e what needs to be set to what?)
        > Alternatively, is there such an option in other software doing simulations?
        >
        > regards & thank you
        > Frank
        >
        >
        >
        >
        > <><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><>
        >
        >
        >
        >
        >
        > > Dear all,
        > >
        > > can someone help me with a a copule of related questions?
        > > A possible approach to incorporate secondary data for estimation is to
        > fit a
        > > "traditional" statistical model using the variable of interest as the
        > > dependent one, to estimate the trend, and to add to it interpolatated
        > > residuals.
        > >
        > > 1)
        > > Is this what Kriging with a variable mean is in GSLIB ?
        > >
        > > if this is the case, does it mean that the variogram to be supplied to
        > > kriging and simulation programs is that of the residuals?
        > >
        > > 2)
        > > Would it be sensible to adopt the approach described above using
        > simulation
        > > (rather than kriging)of residuals? My concern is that one might risk to
        > come
        > > up with realisations in which some locations have negative (and in my
        > case
        > > non sensical) estimates of the variable in question (i.e. how would one
        > deal
        > > with this)?
        > >
        > > thanks very much for your help
        > >
        > > Frank
        >
        >
        > Hi,
        >
        > The answer is yes to your 1st question and I have never
        > experienced problems with simulation using local mean.
        > If it's performed in the normal space, the backtransform
        > would allow one to restrict tthe range of simulated values
        > to positive values.
        >
        > Regards,
        >
        > Pierre Goovaerts
        > ---------------
        > Dear Frank,
        >
        > Regarding your question to GEOSTATS, see my webpage publications for the
        > report on PATTERN-PLUS, which basically combined GLS regression with
        > kriged residuals. I guess the question of simulation vs. kriging depends
        > on whether you want one most-likely answer at each location or a full
        > distribution of possible answers.
        >
        > Yetta
        >
        > _________________________________________________________________
        > It's fast, it's easy and it's free. Get MSN Messenger today!
        > http://www.msn.co.uk/messenger
        >
        >
        > --
        > * To post a message to the list, send it to ai-geostats@...
        > * As a general service to the users, please remember to post a summary of any useful responses to your questions.
        > * To unsubscribe, send an email to majordomo@... with no subject and "unsubscribe ai-geostats" followed by "end" on the next line in the message body. DO NOT SEND Subscribe/Unsubscribe requests to the list
        > * Support to the list is provided at http://www.ai-geostats.org
        >


        --
        * To post a message to the list, send it to ai-geostats@...
        * As a general service to the users, please remember to post a summary of any useful responses to your questions.
        * To unsubscribe, send an email to majordomo@... with no subject and "unsubscribe ai-geostats" followed by "end" on the next line in the message body. DO NOT SEND Subscribe/Unsubscribe requests to the list
        * Support to the list is provided at http://www.ai-geostats.org
      • Edzer J. Pebesma
        ... Yes, http://www.gstat.org -- Edzer -- * To post a message to the list, send it to ai-geostats@unil.ch * As a general service to the users, please remember
        Message 3 of 3 , Jun 24, 2003
        • 0 Attachment
          geogis 123 wrote:

          >
          > Alternatively, is there such an option in other software doing
          > simulations?

          Yes, http://www.gstat.org
          --
          Edzer


          --
          * To post a message to the list, send it to ai-geostats@...
          * As a general service to the users, please remember to post a summary of any useful responses to your questions.
          * To unsubscribe, send an email to majordomo@... with no subject and "unsubscribe ai-geostats" followed by "end" on the next line in the message body. DO NOT SEND Subscribe/Unsubscribe requests to the list
          * Support to the list is provided at http://www.ai-geostats.org
        Your message has been successfully submitted and would be delivered to recipients shortly.