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

AI-GEOSTATS: (no subject)

Expand Messages
  • Ar�. G�r. Fevzi AKBA�
    dear all, I new in geostatistisc and using gs 5.1 sofware. My question is especially about this sofware. When modelling anisotropic variogram I have
    Message 1 of 4 , May 6, 2002
    View Source
    • 0 Attachment
      dear all,
      I new in geostatistisc and using gs 5.1 sofware.
      My question is especially about this sofware. When
      modelling anisotropic variogram I have problems.

      The software only modelling geometric anisotpoy and
      gives major and minor range. Sill and nugget values is
      same. When I change the sill and range visual fits
      getting better but RSS values getting bigger.

      As I understand using this software modelling
      directional variogram is impossible.
      Thank you very much for your interest
      fevzi akbas

      --
      * 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
    • fevzi akbas
      dear all, I modelled a semivariogram and then using this variogram parameter performed cross validation. The computer program gives cross validation results as
      Message 2 of 4 , May 26, 2003
      View Source
      • 0 Attachment
        dear all,

        I modelled a semivariogram and then using this variogram parameter performed cross validation.

        The computer program gives cross validation results as correlation coefficient between estimated and actual values. this value as I know must be close to one.

        Is there any value for correlation coefficient( for example 0,6 or 0,7 ) that can be selected as limit value choosing semivariogram models represent our study area.

        Any literature about this subject will be apreciated.


        --
        * 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
        Hello, I am afraid that your approach of using cross validation results to select the semivariogram model suffers from several limitations. 1. Cross validation
        Message 3 of 4 , May 27, 2003
        View Source
        • 0 Attachment
          Hello,

          I am afraid that your approach of using cross validation results
          to select the semivariogram model suffers from several limitations.
          1. Cross validation results depend on many parameters besides
          the semivariogram model, such as type of kriging, search strategy,
          and it is not obvious which parameter could be responsible
          for poor prediction scores.
          2. The correlation coefficient would be unable to detect a systematic
          bias in your prediction, hence measures like absolute prediction
          errors would be better suited.
          3. Test of hypothesis on correlation coefficients is built on
          the assumption of indepedence of observations which is not the case here.

          At best you could use cross validation scores to compare several
          semivariogram models (instead of focusing on a single one), so use
          this as relative instead of absolute measure. But several studies
          have showed that kriging results are quite robust with respect
          to small changes in the parameters of the semivariogram model.

          Regards,

          Pierre Goovaerts

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

          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, 27 May 2003, fevzi akbas wrote:

          > dear all,
          >
          > I modelled a semivariogram and then using this variogram parameter performed cross validation.
          >
          > The computer program gives cross validation results as correlation coefficient between estimated and actual values. this value as I know must be close to one.
          >
          > Is there any value for correlation coefficient( for example 0,6 or 0,7 ) that can be selected as limit value choosing semivariogram models represent our study area.
          >
          > Any literature about this subject will be apreciated.
          >
          >
          > --
          > * 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
        Your message has been successfully submitted and would be delivered to recipients shortly.