Hi,
after standardizing my data by dividing each covariate by its mean,
the hazard ratio of variables ranges btw ~ 0.5 and 3.55. when i test the PH
assumption by cox.zph, rho ranges btw -0.4235863 and 0.359827.
visualizing the PH assumption by plot, the plot of scaled schoefeld
residuals against transformed time doesn't give a line with slope 0 for
all covariates. note that there are 100 covariates and 300 samples.
Look forward to your reply,
Carol
Check out the all-new Yahoo! Mail beta - Fire up a more powerful email and get things done faster.
|