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

RE: [ai-geostats] Hpw to estimate variance with weighted samples

Expand Messages
  • corinne_danielli@oxy.com
    Thanks for this, Jan. I haven t tried this particular one myself, in EXCEL, but I have found that EXCEL s statistics are unreliable and QC is essential.
    Message 1 of 2 , Oct 7, 2005
    • 0 Attachment
      Message
      Thanks for this, Jan.  I haven't tried this particular one myself, in EXCEL, but I have found that EXCEL's statistics are unreliable and QC is essential.
       
      Corinne Danielli
      -----Original Message-----
      From: JW [mailto:merksmatrix@...]
      Sent: Friday, October 07, 2005 10:55 AM
      To: ai-geostats@...
      Subject: [ai-geostats] Hpw to estimate variance with weighted samples

      Hi Readers,
       
      The correct formula is implemented in several Excel templates that are posted under Documents on ai-geostats. Its veracity can be proved heuristically by replacing all variable weights with unity (1), and watch the weighted average turn into the arithmetic mean, the variance of the weighted average into the Central Limit Theorem, and the degrees of freedom into the positive integer df=n-1.  The formula applies to area-, count-, density-, distance-, length, mass and volume-weighted averages.
       
      The variance of the weighted average derives from the properties of variances. William Volk's Applied Statistics for Engineers, deals exhaustively with the properties of variances.  Amazon.com carries offers preread copies of  Volk's textbook.
       
      Kind regards,
      Jan W Merks  
    Your message has been successfully submitted and would be delivered to recipients shortly.