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

AI-GEOSTATS: Indicator kriging variance?

Expand Messages
  • vanessa stelzenmüller
    Dear list members, I m working at an application of Indicator Kriging in Fisheries and I wonder how to interpret the Indicator kriging variance ? Is it
    Message 1 of 4 , Oct 21, 2002
    • 0 Attachment
      Dear list members,

      I'm working at an application of Indicator Kriging in
      Fisheries and I wonder how to interpret the Indicator
      kriging variance ?
      Is it justanother kind of kriging variance ? or
      Is it usless when talking about a non-parametric
      version of OK?

      I'm looking forward for your comments...

      Best wishes
      Vanessa

      =====
      Vanessa Stelzenmüller,PhD-Student
      ICBM, Dep. Aquatic Ecology
      C.v.O University of Oldenburg
      P.O. Box 2503
      26111 Oldenburg
      <º)))>< <º)))><
      Tel:+49 441-798 3306, Fax:+49 441-798 3701

      __________________________________________________________________

      Gesendet von Yahoo! Mail - http://mail.yahoo.de
      Möchten Sie mit einem Gruß antworten? http://grusskarten.yahoo.de

      --
      * 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
    • Syed Abdul Rahman Shibli
      Practically speaking you re trying to generate a posterior PDF at each location, so the indicator kriging variance might have little or no use. ... useful
      Message 2 of 4 , Oct 21, 2002
      • 0 Attachment
        Practically speaking you're trying to generate
        a posterior PDF at each location, so the indicator
        kriging variance might have little or no use.

        ---- Original message ----
        >Date: Mon, 21 Oct 2002 09:10:46 +0200 (CEST)
        >From: vanessa stelzenmüller <vstelzenmueller@...>
        >Subject: AI-GEOSTATS: Indicator kriging variance?
        >To: ai-geostats@...
        >
        >Dear list members,
        >
        >I'm working at an application of Indicator Kriging in
        >Fisheries and I wonder how to interpret the Indicator
        >kriging variance ?
        >Is it justanother kind of kriging variance ? or
        >Is it usless when talking about a non-parametric
        >version of OK?
        >
        >I'm looking forward for your comments...
        >
        >Best wishes
        >Vanessa
        >
        >=====
        >Vanessa Stelzenmüller,PhD-Student
        >ICBM, Dep. Aquatic Ecology
        >C.v.O University of Oldenburg
        >P.O. Box 2503
        >26111 Oldenburg
        ><º)))>< <º)))><
        >Tel:+49 441-798 3306, Fax:+49 441-798 3701
        >
        >__________________________________________________________________
        >
        >Gesendet von Yahoo! Mail - http://mail.yahoo.de
        >Möchten Sie mit einem Gruß antworten? http://grusskarten.yahoo.de
        >
        >--
        >* 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
        Indicator kriging gives you estimates of the probabilities that you would need to assess the posterior PDF. They are not probabilities, reason why they are
        Message 3 of 4 , Oct 21, 2002
        • 0 Attachment
          Indicator kriging gives you estimates of the probabilities
          that you would need to assess the posterior PDF. They are
          not probabilities, reason why they are outside [0,1] and
          exhibit all kind of order relation violations.

          Syed Abdul Rahman Shibli wrote:
          >
          > Practically speaking you're trying to generate
          > a posterior PDF at each location, so the indicator
          > kriging variance might have little or no use.
          >
          I'd say the emphasis here is on "trying". Maybe the kriging variance
          can indicate whether your IK probability estimates are close,
          or whether you basically do not have a clue about the real probabilities
          of the posterior PDF.

          I know many people throw away (indicator) kriging variances, I just
          do not understand why.

          I would like to see some more opinions on this, on the list.
          --
          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
        • Syed Abdul Rahman Shibli
          ... Part of the motivation for indicator kriging and simulation (at least in the oil and gas domain) is to move away from the normal kriging variance, which is
          Message 4 of 4 , Oct 21, 2002
          • 0 Attachment
            >... Maybe the kriging variance
            >can indicate whether your IK probability estimates are close,
            >or whether you basically do not have a clue about the real probabilities
            >of the posterior PDF.

            Part of the motivation for indicator kriging and simulation
            (at least in the oil and gas domain) is to move away from the
            normal kriging variance, which is just a function of data
            configuration. The other is to estimate the connectivity of extreme
            values, something not too apparent on a single average
            (conventional) variogram. As for the former, there are two kinds
            of uncertainties, for example the probablity that a permeability
            value is above an uncertain threshold (say 20 mdD instead of 30 md),
            and the probability that a permeability value should be within
            a certain class (say anywhere between 30 mD and 70 mD). I guess
            one can generate multiple probablity maps via stochastic simulation
            and generate even more realizations within each probability field,
            but -- at the risk of sounding heretical -- I think this is practically
            more trouble than it's worth.

            Syed

            --
            * 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.