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

Re: AI-GEOSTATS: Outliers problems

Expand Messages
  • Chaosheng Zhang
    Ali, It is always a problem how to handle outliers properly. Generally speaking, it depends on your purpose. The following are just some tips based on my
    Message 1 of 2 , Jan 29, 2003
    • 0 Attachment
      Ali,

      It is always a problem how to handle outliers properly. Generally speaking,
      it depends on your purpose. The following are just some tips based on my
      experience.

      If you would like to estimate the representative mean value, a median, or
      percentiles, or the mean value based on some kind of data transformation
      (towards normality) will be good enough. You may also be aware that the
      samples should be spatially declustered, and they belong to one population.
      You will over-estimate the mean value if the samples are spatially clustered
      in the high value areas.

      If you want to estimate the robust variogram, outliers may be removed in the
      variogram calculation, and be put back in kriging. Using median or other
      methods to estimate the variogram may be considered, but I haven't found it
      effective.

      Outliers are also related to probability distribution. Those which are
      outliers in the raw data may behave well in the transformed data set.
      Backtransformation (for positively skewed data) in kriging may be
      problematic as the estimated values may be lower than they should be in the
      original scale if they were backtransformed directly.

      Hope you find these comments helpful.

      Regards,

      Chaosheng
      =================================================
      Dr. Chaosheng Zhang
      Lecturer in GIS
      Department of Geography
      National University of Ireland
      Galway
      IRELAND

      Tel: +353-91-524411 ext. 2375
      Fax: +353-91-525700
      Email: Chaosheng.Zhang@...
      ChaoshengZhang@...
      Web: http://www.nuigalway.ie/geography/zhang.html
      =================================================


      ----- Original Message -----
      From: "ali qadrouci" <qadrouci_ali@...>
      To: <ai-geostats@...>
      Sent: Wednesday, January 29, 2003 9:31 AM
      Subject: AI-GEOSTATS: Outliers problems


      >
      >
      >
      >
      > Dear listers,
      >
      > I have to apply a cutting on a set of grades coming from underground
      > development or drill holes to eliminate the high values (outliers). As you
      > know these values overestimate the average value and we have a great
      > difference between the grade plant process and the estimated geological
      > grade.
      > Can someone help me for this and give me a method to apply to eliminate
      > these high values.
      >
      > thanks for any advice, Ali
      >
      > _________________________________________________________________
      > MSN Messenger : discutez en direct avec vos amis !
      > http://www.msn.fr/msger/default.asp
      >
      >
      > --
      > * 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.