### Wednesday, January 04, 2006

## Re: st: RE: Nonparametric Multiple Comparison Testing

Chris,

I had the same problem recently and was able solve my problem courtesy of this code written by Joseph Coveney. I have tailored the code to my specific need. It uses Stata's ranksum command to compare the groups using holm's method. Other methods are available, type help _mtest. As it stands this code will only work for four groups due to this line in the code: "forvalues k = `j'/4". Therefore if you have seven groups your code should read "forvalues k = `j'/7".

Here is the link to the original:

http://www.stata.com/statalist/archive/2003-04/msg00273.html

clear set more off infile byte rctime1 byte rctime2 byte rctime3 byte rctime4 using /// http://www.uni-koeln.de/themen/Statistik/data/winer/win_228.txt generate person = _n reshape long rctime, i(person) j(drug) matrix A = J(1, 3, 0) local testno = 0 forvalues i = 1/3 { local j = `i' + 1 forvalues k = `j'/4 { quietly ranksum rctime if (person==`i'|person==`k'), by(person) if r(z)>0 matrix B = (`i', `k', 2*(1-norm(r(z)))) else matrix B = (`i', `k', 2*(norm(r(z)))) if `testno' == 0 { matrix A = B } else { matrix A = A \ B } local rownamesA = "`rownamesA'" + "testpair`++testno' " } } matrix rownames A = `rownamesA' matrix colnames A = onemut othermut unadjustp _mtest adjust A, mtest(holm) pindex(3) append matrix list r(result) exit

Regards, Raphael

* * 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/

Tag: statalist