Dear Allstat-people,
as there was interest in a summary to that thread, and there have been
some follow-ups after I posted the summary, I'll provide those, too
and say thank you's to Barry Nix, Sam Ellis and Stan Alekman.
-------------------------------------------------------
Sam Ellis
-------------------------------------------------------
I had not taken much notice of the original question (because at the
time I did not know the answer) but having seen the summary of replies
I
have remembered something I have seen recently.
I have been looking at the noncentral t distribution in order to
evaluate tolerance limits for the normal distribution. In Johnson,
Kotz
& Balakrishnan, Continuous Univariate Distributions Volume 2, Chapter
31
on the noncentral t distribution there is a discussion on the
estimation
of confidence intervals for the reciprocal of the coefficient of
variation. This refers to a paper (A.T. McKay, Distribution of the
coefficient of variation and the extended t distribution, Journal of
the
Royal Statistical Society, 1932, 95, 695-698) that discusses
approximations to the distribution of the sample coefficient of
variation in relation to the noncentral t distribution. This may be
of
some interest to you.
Sam Ellis
-------------------------------------------------------
Barry Nix
-------------------------------------------------------
Hi
Sorry I missed your original request. I was involved with others
trying to find the distribution of CV when samples are from normal
distributions. We published our results in
Methods of Operational Research 44.
The title was "A criticism of existing external quality assessment
schemes in medical biochemistry". A.B.J.Nix & R.J.Rowlands
Hope this helps.
Barry Nix
-------------------------------------------------------
Stan Alekman
-------------------------------------------------------
Coefficient of variation follows the noncentral F distribution.
Stan Alekman
-------------------------------------------------------
-------------------------------------------------------
Regards
--
-------------------------------------------------------------
Dietrich Alte (Statistician, Dipl.-Stat.)
- SHIP // Project Management -
University of Greifswald - Medical Faculty
Institute of Epidemiology and Social Medicine
Walther-Rathenau-Str. 48, D-17487 Greifswald, Germany
Phone +49(0)3834-867713, fax ++49(0)3834-866684
Email [log in to unmask]
Institute http://www.medizin.uni-greifswald.de/epidem/
-------------------------------------------------------------
|