Search All of the Math Forum:
Views expressed in these public forums are not endorsed by
Drexel University or The Math Forum.


dickreuter
Posts:
56
From:
London
Registered:
10/18/08


Re: p value when using corr2
Posted:
Mar 15, 2009 7:55 AM


On Mar 15, 11:34 am, "i " <ie...@sohu.com> wrote: > Could you introduce more detail? > If I compare two matrices A and B with corr2, then get a r value 0.68, how can I get the p value with ttest or signrank? > > Thank you very much. > > Jin > > NickPick <dickreu...@yahoo.com> wrote in message <7697d1cb4d8a4f28b3dd61592fddb...@o36g2000yqh.googlegroups.com>... > > On Mar 14, 1:54=A0am, "i " <ie...@sohu.com> wrote: > > > Hi, all > > > > Could anyone point out how to get the p value when using corr2 function t= > > o compare two matrices? > > > > Thanks! > > > > Jin > > > see ttest and signrank > >
Type help ttest or help signrank and you get details how to use the two. If I understand you correctly, you'd like to find out whether there is any statistically significant correlation. If the numbers are distributed normally you can use a ttest otherwise you should use a nonparametric test such as the Wilcoxon signrank test.
A = [1 2 44 55 33 54]; B = [12 23 4 55 3 54]; [testresult_ttest,p_ttest]=ttest (A,B) [p_signrank,testresult_signrank]=signrank (A,B)



