Search All of the Math Forum:

Views expressed in these public forums are not endorsed by NCTM or The Math Forum.

Notice: We are no longer accepting new posts, but the forums will continue to be readable.

Topic: p value when using corr2
Replies: 4   Last Post: Mar 15, 2009 8:28 AM

 Messages: [ Previous | Next ]
 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 <7697d1cb-4d8a-4f28-b3dd-61592fddb...@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
non-parametric 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)

Date Subject Author
3/13/09 i
3/14/09 dickreuter
3/15/09 i
3/15/09 dickreuter
3/15/09 i