R help - May 2010 - Comparing the correlations coefficient of two (very) dependent samples

Hello all,

I believe this can be done using bootstrap, but I am wondering if there is
some other way that might be used to tackle this.

#Let's say I have two pairs of samples:
set.seed(100)
s1 <- rnorm(100)
s2 <- s1 + rnorm(100)

x1 <- s1[1:99]
y1 <- s2[1:99]

x2 <- x1
y2 <- s2[2:100]

#And both yield the following two correlations:
cor(x1,y1) # 0.7568969  (cor1)
cor(x2,y2) # -0.2055501 (cor2)

Now for my questions:
1) is cor1 larger then cor2?   (CI for the diff ?)
2) With what P value?
3) What if the values of s1 are not independent ?

I found an older thread discussing such issues:
http://tolstoy.newcastle.edu.au/R/e2/help/06/09/1035.html
But wasn't sure how much this might be relevant to my case.



Thanks for any help,
Tal

----------------Contact
Details:-------------------------------------------------------
Contact me: Tal.Galili@gmail.com |  972-52-7275845
Read me: www.talgalili.com (Hebrew) | www.biostatistics.co.il (Hebrew) |
www.r-statistics.com (English)
----------------------------------------------------------------------------------------------

	[[alternative HTML version deleted]]

Dear Tal,

There are several approaches in doing it (see Steiger, 2003). It
should not be difficult to implement them in R.

Steiger, J.H. (2003). Comparing correlations. In A. Maydeu-Olivares
(Ed.) Psychometrics. A festschrift to Roderick P. McDonald. Mahwah,
NJ:  Lawrence Erlbaum Associates.
http://www.statpower.net/Steiger%20Biblio/Steiger03.PDF

Regards,
Mike
--
---------------------------------------------------------------------
 Mike W.L. Cheung               Phone: (65) 6516-3702
 Department of Psychology       Fax:   (65) 6773-1843
 National University of Singapore
 http://courses.nus.edu.sg/course/psycwlm/internet/
---------------------------------------------------------------------

On Tue, May 4, 2010 at 2:00 AM, Tal Galili <tal.galili at gmail.com>
wrote:
> Hello all,
>
> I believe this can be done using bootstrap, but I am wondering if there is
> some other way that might be used to tackle this.
>
> #Let's say I have two pairs of samples:
> set.seed(100)
> s1 <- rnorm(100)
> s2 <- s1 + rnorm(100)
>
> x1 <- s1[1:99]
> y1 <- s2[1:99]
>
> x2 <- x1
> y2 <- s2[2:100]
>
> #And both yield the following two correlations:
> cor(x1,y1) # 0.7568969 ?(cor1)
> cor(x2,y2) # -0.2055501 (cor2)
>
> Now for my questions:
> 1) is cor1 larger then cor2? ? (CI for the diff ?)
> 2) With what P value?
> 3) What if the values of s1 are not independent ?
>
> I found an older thread discussing such issues:
> http://tolstoy.newcastle.edu.au/R/e2/help/06/09/1035.html
> But wasn't sure how much this might be relevant to my case.
>
>
>
> Thanks for any help,
> Tal
>
> ----------------Contact
> Details:-------------------------------------------------------
> Contact me: Tal.Galili at gmail.com | ?972-52-7275845
> Read me: www.talgalili.com (Hebrew) | www.biostatistics.co.il (Hebrew) |
> www.r-statistics.com (English)
>
----------------------------------------------------------------------------------------------
>
> ? ? ? ?[[alternative HTML version deleted]]
>
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide
http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
>