[R] k-sample Kolmogorov-Smirnov test?

Rolf Turner r.turner at auckland.ac.nz
Mon Sep 15 00:26:46 CEST 2008


On 15/09/2008, at 10:02 AM, Adam D. I. Kramer wrote:

> Maybe you should look a little harder.
>
> help.search("Kolmogorov")
>
>> PLEASE do read the posting guide http://www.R-project.org/posting- 
>> guide.html

Please do read the guy's question!!!

	Hello, I would like to conduct a k-sample K-S test, but cannot find
	reference to its implementation in R. Does anyone have experience  
with this?

It's about a ***k-sample*** K-S test.  As far as I can discern,  
help.search("Kolmogorov")
points one only to ks.test() which (again, as far as I can discern)  
effects only one
or two sample KS tests.

The multi-sample KS test does exist --- I must admit this was a new  
one on me.  (But then,
so many things are!) See e.g. JASA vol. 68, No. 344, pp. 994--997  
``Tables of Critical Values
for a k-sample Kolmogorov-Smirnov Test Statistic'' by Edward H. Wolf  
and Joseph I. Naus.

		cheers,

			Rolf Turner

P.S. RSiteSearch("Kolmogorov k-sample") turned up nothing useful.

			R. T.

######################################################################
Attention:\ This e-mail message is privileged and confid...{{dropped:9}}



More information about the R-help mailing list