Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

Re: st: test of two spearman correlations?


From   "Wagner, Joseph" <WagnerJ@edc.pitt.edu>
To   "STATAList (statalist@hsphsun2.harvard.edu)" <statalist@hsphsun2.harvard.edu>
Subject   Re: st: test of two spearman correlations?
Date   Tue, 17 Feb 2004 16:36:16 -0500

Thank you, -suest- worked perfectly.  I was wondering though, would it would
be appropriate in a situation where I have done two -nptrend-s to test if
the two coefficients vary between groups?

The variables are race (1=Caucasian American or CA,  2=African American or
AA), vload (viral load, continuous), and altcat (categorical levels (3) of
ALT)

In my case, viral load is not normal and no transformation even approaches
normal.

Would the following be appropriate:

nptrend vload if race==1, by(altcat)
nptrend vload if race==2, by(altcat)

egen raltca = rank(age)   if race==1
egen rvica  = rank(vload) if race==1
egen raltaa = rank(age)   if race==2
egen rviaa  = rank(vload) if race==2

regress rvica raltca, score(scoca)
estimates store ca
regress rviaa raltaa, score(scoaa)
estimates store aa

suest aa ca
test [ca_mean]raltca = [aa_mean]raltaa



>You might be able to use -bootstrap- (or -jknife-) to obtain confidence
>intervals of the correlation coefficients (or of their difference), and
there
>are probably other more direct methods, too, but would something like the
>following be worth considering?
>
*
*   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