R help - Aug 2011 - some questions about stepwise cox regression model and heterogeneity in survival analysis

Hello, all users here:

      Recently i am doing a project of survival analysis. I collect the
characteristics of patients and have got some factors which are related with
the cancer.

      When i come to overcome relations between genotypes (Snps) and
survival time via stratified analysis, i come across the problem of how to
use coxph(survival package) to performe multivariate cox regression analysis
with adjustment for some potential confounders. I have not found any
suggestions of the survival package manual.

      And when i want to calculate the heterogeneity between genotypes and
other stratified factors, i can not find something command like this (only
survdiff for log-rank test, but no command for heterogeneity test). 

      At last, stepwise cox regression need to use a significance level of
0.05 for entering and 0.10 for removing the explanatory variables. First how
can i do stepwise cox regression analysis using survival package. And how
can i limit  the significance level?


    Could anyone here offer me the solution? Thank a lot, it will be
appreciate of your kindly answers!

--
View this message in context:
http://r.789695.n4.nabble.com/some-questions-about-stepwise-cox-regression-model-and-heterogeneity-in-survival-analysis-tp3755591p3755591.html
Sent from the R help mailing list archive at Nabble.com.