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

GEOSTATS: Trend Surface Analysis software

Expand Messages
  • Burnett, Mark
    Hi All Could someone please advise me on a decent share/free ware package that has the ability to run various forms of trend surface analysis on large data
    Message 1 of 5 , Oct 24, 2000
      Hi All

      Could someone please advise me on a decent share/free ware package that has
      the ability to run various forms of trend surface analysis on large data
      sets (100 000+) e.g. residual, polynomial etc. (URL site for download would
      be appreciated), and contour the data set afterwards.

      The current software that I am using does not have this functionality built
      in and I want to run a number of checks on various parameters that I have in
      the data set.

      Thank you

      Mark Burnett
      Evaluation Manager

      --
      *To post a message to the list, send it to ai-geostats@....
      *As a general service to list users, please remember to post a summary
      of any useful responses to your questions.
      *To unsubscribe, send email to majordomo@... with no subject and
      "unsubscribe ai-geostats" in the message body.
      DO NOT SEND Subscribe/Unsubscribe requests to the list!
    • Mark Hall
      Have you checked out the R package available from www.lib.stat.cmu.edu/R and its various add-on packages. All free! Best, Mark Hall ... From: Burnett, Mark
      Message 2 of 5 , Oct 24, 2000
        Have you checked out the R package available from www.lib.stat.cmu.edu/R and
        its various add-on packages. All free!

        Best, Mark Hall


        ----- Original Message -----
        From: "Burnett, Mark" <MBurnett@...>
        To: <ai-geostats@...>
        Sent: Tuesday, October 24, 2000 10:31 PM
        Subject: GEOSTATS: Trend Surface Analysis software


        > Hi All
        >
        > Could someone please advise me on a decent share/free ware package that
        has
        > the ability to run various forms of trend surface analysis on large data
        > sets (100 000+) e.g. residual, polynomial etc. (URL site for download
        would
        > be appreciated), and contour the data set afterwards.
        >
        > The current software that I am using does not have this functionality
        built
        > in and I want to run a number of checks on various parameters that I have
        in
        > the data set.
        >
        > Thank you
        >
        > Mark Burnett
        > Evaluation Manager
        >
        > --
        > *To post a message to the list, send it to ai-geostats@....
        > *As a general service to list users, please remember to post a summary
        > of any useful responses to your questions.
        > *To unsubscribe, send email to majordomo@... with no subject and
        > "unsubscribe ai-geostats" in the message body.
        > DO NOT SEND Subscribe/Unsubscribe requests to the list!
        >
        >

        --
        *To post a message to the list, send it to ai-geostats@....
        *As a general service to list users, please remember to post a summary
        of any useful responses to your questions.
        *To unsubscribe, send email to majordomo@... with no subject and
        "unsubscribe ai-geostats" in the message body.
        DO NOT SEND Subscribe/Unsubscribe requests to the list!
      • Mark Hall
        Sorry folks about messing up the address for Stat Lib, it is: http://lib.stat.cmu.edu/ Best, Mark Hall ... From: To: Mark Hall
        Message 3 of 5 , Oct 25, 2000
          Sorry folks about messing up the address for Stat Lib, it is:

          http://lib.stat.cmu.edu/

          Best, Mark Hall


          ----- Original Message -----
          From: <abedini@...>
          To: "Mark Hall" <hall@...>
          Sent: Thursday, October 26, 2000 1:55 AM
          Subject: Re: GEOSTATS: Trend Surface Analysis software


          >
          > Dear Mark
          >
          > I tried to connect to the URl which you provided receiving the following
          > reply. Are you sure the address is correct?
          >
          > Thanks
          > Abedini
          >
          > While trying to retrieve the URL: http://www.lib.stat.cmu.edu/
          >
          > The following error was encountered:
          >
          > Unable to determine IP address from host name for www.lib.stat.cmu.edu
          > The dnsserver returned:
          >
          > Name Error: The domain name does not exist.
          > This means that:
          >
          > The cache was not able to resolve the hostname presented in the URL.
          > Check if the address is correct.
          >
          > Your cache administrator is root.
          >
          >
          >
          > --------------------------------------------------------------------------
          ------
          > Generated Wed, 25 Oct 2000 16:51:46 GMT by proxy.shirazu.ac.ir
          > (Squid/2.3.STABLE3)
          >
          >
          >

          --
          *To post a message to the list, send it to ai-geostats@....
          *As a general service to list users, please remember to post a summary
          of any useful responses to your questions.
          *To unsubscribe, send email to majordomo@... with no subject and
          "unsubscribe ai-geostats" in the message body.
          DO NOT SEND Subscribe/Unsubscribe requests to the list!
        • Roger Bivand
          ... The spatial package can be used (Venables & Ripley, MASS 3rd ... Min. 1st Qu. Median Mean 3rd Qu. Max. -4.119 5.855 7.915 7.965 10.040
          Message 4 of 5 , Oct 26, 2000
            On Wed, 25 Oct 2000, Mark Hall wrote:

            > Have you checked out the R package available from www.lib.stat.cmu.edu/R and
            > its various add-on packages. All free!
            >
            > Best, Mark Hall

            The "spatial" package can be used (Venables & Ripley, MASS 3rd
            edition; package in the VR bundle on CRAN:

            > x <- runif(100000)
            > y <- runif(100000)
            > z <- 5*x + 3*y + 0.5*x*y + 1*x*x + 1.5*y*y + rnorm(100000, 3, 2)
            > summary(z)
            Min. 1st Qu. Median Mean 3rd Qu. Max.
            -4.119 5.855 7.915 7.965 10.040 21.110
            > library(spatial)
            > tr2 <- surf.ls(2, x, y, z)
            > str(tr2)
            List of 11
            $ x : num [1:100000] 0.111 0.793 0.581 0.868 0.254 ...
            $ y : num [1:100000] 0.7505 0.7083 0.3539 0.0571 0.1982 ...
            $ z : num [1:100000] 2.12 12.15 8.04 5.71 -1.65 ...
            $ np : num 2
            $ f : num [1:600000] 1 1 1 1 1 1 1 1 1 1 ...
            $ r : num [1:21] -3.16e+02 -3.67e-01 -1.82e+02 -1.05e+02 -7.89e-03 ...
            $ beta: num [1:6] 7.753 3.101 0.273 2.376 0.127 ...
            $ wz : num [1:100000] -4.620 1.400 0.449 -2.569 -6.682 ...
            $ rx : num [1:2] 6.04e-06 1.00e-00
            $ ry : num [1:2] 2.71e-05 1.00e-00
            $ call: language surf.ls(np = 2, x = x, y = y, z = z)
            - attr(*, "class")= chr "trls"
            > trsurf2 <- trmat(tr2, 0, 1, 0, 1, 100)
            > image(trsurf2)
            > contour(trsurf2, add=T)
            > tr3 <- surf.ls(3, x, y, z)
            > image(trsurf2)
            > contour(trsurf3, add=T)

            However, R does all its work in memory, so I needed a heap size of 32M to
            complete the quadratic trend, and over 40M for the cubic. If this is OK,
            then the surf.ls() function and trmat() to generate predictions will do
            what you need. If you need predictions for other points than a grid over
            your study area, you need to look at the way trmat() works - ask if you
            need help.

            Best wishes,

            Roger

            >
            > ----- Original Message -----
            > From: "Burnett, Mark" <MBurnett@...>
            > To: <ai-geostats@...>
            > Sent: Tuesday, October 24, 2000 10:31 PM
            > Subject: GEOSTATS: Trend Surface Analysis software
            >
            >
            > > Hi All
            > >
            > > Could someone please advise me on a decent share/free ware package that
            > has
            > > the ability to run various forms of trend surface analysis on large data
            > > sets (100 000+) e.g. residual, polynomial etc. (URL site for download
            > would
            > > be appreciated), and contour the data set afterwards.
            > >
            > > The current software that I am using does not have this functionality
            > built
            > > in and I want to run a number of checks on various parameters that I have
            > in
            > > the data set.
            > >
            > > Thank you
            > >
            > > Mark Burnett
            > > Evaluation Manager
            > >
            >

            --
            Roger Bivand
            Economic Geography Section, Department of Economics, Norwegian School of
            Economics and Business Administration, Breiviksveien 40, N-5045 Bergen,
            Norway. voice: +47 55 95 93 55; fax +47 55 95 93 93
            e-mail: Roger.Bivand@...
            and: Department of Geography and Regional Development, University of
            Gdansk, al. Mar. J. Pilsudskiego 46, PL-81 378 Gdynia, Poland.

            --
            *To post a message to the list, send it to ai-geostats@....
            *As a general service to list users, please remember to post a summary
            of any useful responses to your questions.
            *To unsubscribe, send email to majordomo@... with no subject and
            "unsubscribe ai-geostats" in the message body.
            DO NOT SEND Subscribe/Unsubscribe requests to the list!
          • Isobel Clark
            If you guys are that desperate I published a program in Computers and Geosciences in 1977 (I think). The reference will be in
            Message 5 of 5 , Oct 26, 2000
              If you guys are that desperate I published a program
              in Computers and Geosciences in 1977 (I think). The
              reference will be in
              http://uk.geocities.com/drisobelclark/resume/publications.html

              It was called SNARK and was for 3-d data.

              Trend Surface Analysis is basically just ordinary
              least squares. Any stat package which does least
              squares regression should do all the relevant tests as
              well.

              The `theory' of PTSA and Analysis of Variance testing
              is in Practical Geostatistics 2000, Chapter 7.

              If you are willing (or allowed) to share your data
              with the world, contact me and I'll turn it into a
              valid file for the free 'book' software

              http://uk.geocities.com/drisobelclark/PG2000_demo.html

              Isobel Clark
              +27 11 726 3729

              ____________________________________________________________
              Do You Yahoo!?
              Get your free @... address at http://mail.yahoo.co.uk
              or your free @... address at http://mail.yahoo.ie
              --
              *To post a message to the list, send it to ai-geostats@....
              *As a general service to list users, please remember to post a summary
              of any useful responses to your questions.
              *To unsubscribe, send email to majordomo@... with no subject and
              "unsubscribe ai-geostats" in the message body.
              DO NOT SEND Subscribe/Unsubscribe requests to the list!
            Your message has been successfully submitted and would be delivered to recipients shortly.