Final edits SRF
catavallejos opened this issue · 0 comments
catavallejos commented
- Output of
plot.competing.risk()needs further explanation. Overall estimates ignoring covariates? Akin to a Aalen-Johansen estimator for the CIF?
No, these are OOB estimates and equations for them are given on the paper. They do depend on x
- Whilst the first type of splitting rule focuses on covariate effects through the CS hazard, you can still obtain CIF estimates (please correct me if I am wrong). I think you can get them using
rsf.lr$cif. It would be good to show those (rather than the CS hazard) as it's more consistent with other vignettes. - #16
I added predictions of the CIF at t=5. Note that we cannot obtain predictions at arbirary time points