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

Re: AI-GEOSTATS: Block size - estimation variance - resource category.

Expand Messages
  • Isobel Clark
    ... Hi Bill It is possible for the kriging variance to be higher than the total sill of the semi-variogram. This total sill is (theoretically) equal to the
    Message 1 of 1 , Mar 16, 2001
    • 0 Attachment
      --- William Thayer <thayer@...> wrote: >
      Isobel:
      > Would you mind expanding a little on your earlier
      > reply? In particular,
      > what do you mean by (1) "kriging variance less than
      > ordinary
      > >statistical sample variance", (2) "not measured but
      > within the range
      > >of influence of at least 4 samples" and (3)
      > "indicated is within range but
      > >the local average is a better estimator than the
      > kriging"?
      >
      > Thanks in advance,
      > Bill

      Hi Bill

      It is possible for the kriging variance to be higher
      than the total sill of the semi-variogram. This total
      sill is (theoretically) equal to the Normal population
      variance if your data is (a) stationary and (b)
      Normal. For lognormals use logarithms. Or transform
      data to Normal scores before calculating
      semi-variogram.

      For example: if you estimate a point location from a
      single sample just below the range of influence away,
      the kriging variance is twice the total sill (unless
      you follow the Stanford school, in which case it is
      twice the total sill minus twice the nugget effect).

      Now, if the kriging variance is higher than the
      'sample' variance, it means that the population mean
      (if you knew it) would be a better estimator than the
      local kriging estimate.

      So even if you have samples within the range of
      influence, you could still get an estimate which has
      worse confidence than the regional average. This I
      call "indicated". I'm sure it is there but I can't put
      a local value on it.

      If the kriging variance is less than the total sill
      (suitably modified for non-point support), then one
      can assign a 'local' value which is better than the
      regional average. I consider this "measured".

      Anything outside the range of influence is speculative
      and the province of the geologist. I am a mining
      engineer and don't do "inferred".

      Does this help?
      Isobel
      http://uk.geocities.com/drisobelclark

      PS: I have fond memories of Syracuse. Spent the summer
      of '76 as visiting prof in geology.


      ____________________________________________________________
      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 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.