Deat Tipsters,
A number of people, particularly Karl and John have pointed out that
s is not an unbiased estimator of sigma. My post claiming such was
hasty - I was thinking of variances, where sSqu is an unbiased
estimator of sigmasqu if n-1 is used. The expectation algebra shows
this. However, Karl's fine example shows how this conclusion does
not follow for s - although using n-1 still gives a better estimate
than using n.
Karl was excused stats. classes for the rest of the term with good
reason!
Stuart
___________________________________________________
Stuart J. McKelvie, Ph.D., Phone: (819)822-9600
Department of Psychology, Extension 2402
Bishop's University, Fax: (819)822-9661
3 Route 108 East,
Lennoxville, e-mail: [EMAIL PROTECTED]
Quebec J1M 1Z7,
Canada.
Bishop's University Psychology Department Web Page:
http://www.ubishops.ca/ccc/div/soc/psy
___________________________________________________
