[datatable-help] Analyzing reweighted survival data that is correlated

Stanton Hudja Stanton.Hudja at gmail.com
Tue Mar 27 18:20:28 CEST 2018


I'm currently analyzing data from an experiment where I have multiple
observations from each subject. I have re-weighted the data to better
approximate the data generating process. I have also created adjusted
survival curves using the Kaplan-Meier estimator for a control and treatment
group but I am unsure as the best way to test for significance.

The best approach I have thought of so far is to take the means of each
subject under the weighted KM and run hypothesis tests on that. The log-rank
test would have an issue with the correlated data. The cox ph doesn't seem
to have proportional hazards, although I don't think the effect of the
non-proportional hazards matters much with this data.

I would appreciate any advice on any of these tests.

Thanks,
Stanton



--
Sent from: http://r.789695.n4.nabble.com/datatable-help-f2315188.html


More information about the datatable-help mailing list