Statalist


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

Re: st: kolmogorov-Smirmov Output


From   Neil Shephard <[email protected]>
To   [email protected]
Subject   Re: st: kolmogorov-Smirmov Output
Date   Tue, 13 Jan 2009 10:33:04 +0000

On Tue, Jan 13, 2009 at 8:51 AM,  <[email protected]> wrote:
> Hello,
>
> would someone guide me on how to interpret the output of a two-sided
> kolmogorov-Smirnov test?
> In order to conclude for the equality or not of the distributions, what
> p-value I have to see? As you know, if I perform a test for the equality of
> two distributions, I have three rows where the last one is the combined K-S.
> My question is if I have to see only this last row (p-value) in ordwer to
> drow a conclusion.

The p-value tells you the probability of obtaining the observerd
result by chance alone (the same as with all other p-values).   The
threshold for declaring statistical significance should be determined
by the framework of the study you are doing.

The other statistics (i.e. the 'D' column in the output) is a mesaure
of the difference between the two distributions.  These give an
indication of the size of the difference between the two groups.  The
first one tells you if your first group is smallest, the second if
your second is smallest, whilst the third is a two-way test and tells
you if there is a difference (irrespective of the direction of the
difference).

An excellent reference on Kolmogorov-Smirnov is chapter 6 of Connover
W.J. (1999) Practical Non-Parametric Statistics 3rd Edition. John
Wiley & Sons

Neil
-- 
"We should make things as simple as possible, but not simpler."  -
Anon (not Albert Einstein)

Email - [email protected]
Website - http://slack.ser.man.ac.uk/
Photos - http://www.flickr.com/photos/slackline/
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/



© Copyright 1996–2024 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index