Douglas asked:
> ...
> Seondly, I have several treatment groups. To evaluate the
> significance
> of the differences in lifespans between the groups I performed a
> logrank test. Now I would like to perform a series of pairwise
> comparisons between the groups. What is the appropriate procedure to
> protect against multiple comparisons (is there something like
> Bonferroni or Scheffe that should follow the logrank)?
Maybe -_mtest- will give you what you want. Assume X1, X2, X3, ... are
the variables identifying the pairs of groups, then type something like:
foreach X of var X1 X2 X3 ... {
sts test `X'
matrix pvalues=nullmat(pvalues)\ 1-chi2(`r(df)',`r(chi2)')
}
_mtest adjust pvalues, mtest(bonferroni)
matrix list r(result)
matrix drop pvalues
(the unadjusted p-values of the single tests are saved to the matrix
"pvalues"; the matrix "r(result)" contains the bonferroni-ajusted
p-values)
ben
*
* 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/
