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

GEOSTATS: Non-parametrics

Expand Messages
  • Eduardo R. Videla
    Hi, I would like some advise on any available software for doing non-parametric statistics (or better geostatistics, Henley 1981) on large, very large
    Message 1 of 2 , Mar 10, 1997

      Hi,

      I would like some advise on any available software for doing non-parametric statistics
      (or better geostatistics, Henley 1981) on large, very large datasets.

      These datasets comprises multivariate spatial data, in 3D space from severeal gold mines
      in a region in the form

      X, Y ,Z, Au, As, Sb, Rock-type

      I would like to do an analysis of spatial autocorrelation within mines, between mines and an analysis of their cross-correlation, i.e., to test the hypothesis that one of the datasets is related or not to the rest.

      I realise that the lognormal transformation is very close to the observed distribution, mainly in gold, but it still yields a lot of "almost lognormal", thats why I want to use distribution-free statistics and test of hypotheses.

      The problem is the size of the database, some times many thousands of observations and the ability of the softwares to handle it

      Suggestions welcomed. Many thanks   

      ***********************************************
      Eduardo Videla
      Key Centre-Geology Dept.-Un. of Western Australia
      E-mail: evidela@...

    • Chaosheng Zhang
      Hi, Eduardo Videla, Perhaps you may try power transformation for the datasets, such as Box-Cox. Chaosheng Zhang ... Dr. Chaosheng Zhang Division of
      Message 2 of 2 , Mar 10, 1997
        Hi, Eduardo Videla,

        Perhaps you may try power transformation for the datasets, such as Box-Cox.

        Chaosheng Zhang


        >Hi,
        >
        >I would like some advise on any available software for doing non-parametric
        >statistics
        >(or better geostatistics, Henley 1981) on large, very large datasets.
        >
        >These datasets comprises multivariate spatial data, in 3D space from
        >severeal gold mines
        >in a region in the form
        >
        >X, Y ,Z, Au, As, Sb, Rock-type
        >
        >I would like to do an analysis of spatial autocorrelation within mines,
        >between mines and an analysis of their cross-correlation, i.e., to test the
        >hypothesis that one of the datasets is related or not to the rest.
        >
        >I realise that the lognormal transformation is very close to the observed
        >distribution, mainly in gold, but it still yields a lot of "almost
        >lognormal", thats why I want to use distribution-free statistics and test
        >of hypotheses.
        >
        >The problem is the size of the database, some times many thousands of
        >observations and the ability of the softwares to handle it
        >
        >Suggestions welcomed. Many thanks
        >
        >***********************************************
        >Eduardo Videla
        >Key Centre-Geology Dept.-Un. of Western Australia
        >E-mail: evidela@...
        >
        Dr. Chaosheng Zhang
        Division of Geochemistry
        Geological Survey of Sweden, SGU
        P.O. Box 670
        S-751 28 Uppsalla
        SWEDEN
        Tel: +46 18 179290
        Fax: +46 18 179210
        E-mail: chaosheng.zhang@...

        --
        *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.