Sts test across sample and sub-sample

[Steffen Haldrup]

Dear all.

I want to test the equality of two survivor functions, however, it should be across the the full sample and only a part of the sample. I can't really find a solution for doing that, as the strata function takes part in two sub-groups and not the full sample vs. part of the sample.

[Clyde Schechter]

In general, there are no statistical tests that directly contrast a full sample with a subset of that sample. But logically, contrasting a subset with the complementary subset is equivalent to contrasting a subset with the whole. So just do it as contrasts between your part of the sample and the rest of the sample.

[Steve Samuels]

If you have a survey sample, don't forget to svyset the data too. You won't be able to use sts test (it doesn't accept a svy prefix). but you can use svy: stcox to make the comparisons. See Example 3 and the following technical note in the Survey Manual entry for "svy estimation "