calculation of standard deviation is not as help document says

S

samemistake

in the help manual of STDEV, it is said that the calculation uses the
"unbiased" formula, but in fact the actual calculation uses the biased one,
i.e. sqrt( sum(xi-xavg)/N), instead of the intended: sqrt(
sum(xi-xavg)/(N-1)).
am i right? if so, please correct the help document, to avoid the confusion.

----------------
This post is a suggestion for Microsoft, and Microsoft responds to the
suggestions with the most votes. To vote for this suggestion, click the "I
Agree" button in the message pane. If you do not see the button, follow this
link to open the suggestion in the Microsoft Web-based Newsreader and then
click "I Agree" in the message pane.

http://www.microsoft.com/office/com...0-73114b796ca6&dg=microsoft.public.excel.misc
 
D

David Biddulph

samemistake said:
in the help manual of STDEV, it is said that the calculation uses the
"unbiased" formula, but in fact the actual calculation uses the biased
one,
i.e. sqrt( sum(xi-xavg)/N), instead of the intended: sqrt(
sum(xi-xavg)/(N-1)).
am i right? if so, please correct the help document, to avoid the
confusion.
Click to expand...

No, You're wrong, I think.

It looks to me as if STDEV uses N-1 and STDEVP uses N, which is what the
help says. I'm using Excel 2003.

What numbers have you used to convince yourself that it's wrong? Which
version of Excel?
 
Top