Chapter 5 feedback

[tombisho]

Here there are the two options to either generate the survival object separately or in line. It would be good to know the reasons for using each approach. There could be more explanation about strata.

I think also the assign function needs to be included here as another option. The motivation for creating the model on the server side could be borrowed from the ds.glmassign functions? (at a guess it is in case you want to use the model for prediction?).

It would be good to understand how cox_model_full is different to the standard output from survival (I think some things are supressed?). And is it possible to run summary(cox_model_full) on the client side?

I think it would be nice to show the results from these steps, so can we work together to find out why the code won't run against opal-sandbox?