Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

Re: st: permutations


From   Christopher Intemann <intemann@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: permutations
Date   Mon, 17 Mar 2008 18:19:42 +0100

Am 17.03.2008 um 17:49 schrieb David Airey:

The p value is simply p=c/n in the tabled results. It is 0.002, and 0.000. Increase your number of permutations if you want something more exact than 2/1000, or 0/1000. Try 10,000 permutations, instead of 1000. You should try to report an exact probability. This is especially important if you are running many SNPs and later want to apply a multiple test correction procedure. There, 1x10-6 or 1x10-7 matters a lot.

The seed is necessary for replication. If you don't set the seed you will get chance variation in your results, because permute will be using different random numbers; setting the seed makes permute use the same random numbers. It will not affect your statistical conclusions. Most people will seed the random number generator.

Dave,
thank you very, very much.
Your comments where really helpful!
Regards,
 Chris

*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/



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