Dear survival analysis experts, can anybody give me some information about frailty models in S-Plus 2000? Is it possible to fit univariate frailty models in S-PLUS? Evaluating the hazards of two treatment groups I detected a cross-over phenomenom. I think it results from a selection effect of the frailty type due to a heterogenous subpopulation. The heterogeneity can result in a "pulled down" hazard rate with increasing time. In my data set the frailty seems to influence the relative risk which is a decreasing function of time and a cross-over takes place when the high risk group becomes the low risk group. Does anybody know how to test this hypothesis with S-Plus using a univariate frailty model. I tried it with the function "frailty.gamma" assuming a Gamma frailty distribution. I didn't know how to use the options of this function, especially the df option. If I have only one survival time per ind. (univariate data) and I use the function without option parameters I have no df left for a test of the frailty term. How to set the df and the other parameters to solve my problem?: I want to estimate the variance parameter of the frailty distribution to assess heterogenity. The idea is the following: If homogeneity is present the survival times in a specific covariate pattern (subpopulation) should be distributed only by random error-the frailty variance should be small. If there is heterogeneity in a specific subpopulation due to some more FRAIL individuals (with a large frailty term) which die earlier there should be a frailty variance significantly different from zero. Thanks in advance Bernd Genser =========================================================================== This list is open to everyone. Occasionally, less thoughtful people send inappropriate messages. Please DO NOT COMPLAIN TO THE POSTMASTER about these messages because the postmaster has no way of controlling them, and excessive complaints will result in termination of the list. For information about this list, including information about the problem of inappropriate messages and information about how to unsubscribe, please see the web page at http://jse.stat.ncsu.edu/ ===========================================================================
