Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down on April 23, and its replacement, statalist.org is already up and running.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: Using Kolmogorov-Smirnov test (ksmirnov)


From   Nick Cox <njcoxstata@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: Using Kolmogorov-Smirnov test (ksmirnov)
Date   Thu, 21 Feb 2013 18:03:14 +0000

Quantile-quantile plots show more information and make the question of
adjusting for location and scale the side-issue you want it to be. In
your situation you expect a linear configuration. If you standardise
too,  you expect that to be equality. -qqplot- beckons. If someone in
power over you expects inferential mumbo-jumbo too, then get an idea
of variability from e.g. bootstrapping.

When it's sensitive, K-Smirnov is sensitive to differences in the
middle of distributions. For most practical statistics, you need to
worry more about the tails.

Nick

On Thu, Feb 21, 2013 at 5:36 PM, Zachary Neal <zpneal@gmail.com> wrote:
> I would like to determine whether two samples have the same
> distribution form, despite having means and variances that are known
> (and expected) to be different.  Before performing a
> Kolmogorov-Smirnov test, is it necessary to transform the two samples
> so they have a common mean? A common variance?  I'm inclined to use
> z-scores so that any difference detected by ksmirnov is attributable
> to a difference in form, and not to a difference in mean or variance,
> but I'm not sure this is right.
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/


© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index