0

I have done a multivariate Cox regression in R. The model fits to my data very well.

Now, I would like to use my model and predict the survival probabilities of new observations. I am unclear how to do this with a Cox model. Something similar to what can be done using a logistic regression,to plugin the values of new observations and multiply them with betas and calculate the probabilities.

How can this be done in a Cox model?

1 Answers1

1

If you use the survfit function in the survival package on a coxph object then it will provide the predicted survival function (at the specified set of covariates). You can print or plot the entire function. Also from the survfit.object you can calculate the proportion surviving at time points of interest.

Greg Snow
  • 46,563
  • 2
  • 90
  • 159