On 26 Jan 2000 16:16:21 -0800, [EMAIL PROTECTED] (Simon, Steve, PhD)
wrote:
> I forgot to give the full reference for Hollander and Wolfe.
>
> Hollander, Myles and Wolfe, Douglas A. (1999) Nonparametric Statistical
> Methods, Second Edition. New York NY: John Wiley and sons, Inc. ISBN:
> 0-471-19045-4.
>
> For the beginning student. A comprehensive guide to test statistics that do
> not require distributional assumptions (like normality).
- oh, I do think the beginning student should be warned STRONGLY that
the distribution assumptions are *nearly* as strong. That is case
when the first step is rank-transformation, which is very often where
they start.
You still can't stuff your numbers into a computer program and
automatically be told what makes sense about the data. There is
generally an assumption that groups (say) have distributions of the
similar form, and that is most of the *assumption* required if you
want to do something that is parametric and more flexible.
Sorry; one of the things I most dislike about "nonparametric testing"
is the over-selling of it -- "reduced distributional assumptions"
would be a more measured comment.
--
Rich Ulrich, [EMAIL PROTECTED]
http://www.pitt.edu/~wpilib/index.html
===========================================================================
This list is open to everyone. Occasionally, people lacking respect
for other members of the list send messages that are inappropriate
or unrelated to the list's discussion topics. Please just delete the
offensive email.
For information concerning the list, please see the following web page:
http://jse.stat.ncsu.edu/
===========================================================================
