Flexible parametric competing risks models

[mattwarkentin]

Hi @chjackson,

Do you have any thoughts about whether this package would ever add support for being able to use flexible parametric models to directly model the cause-specific cumulative incidences? In particular, I'm wondering if the methods described in this paper (available in Stata via stmp2cr) would be a good candidate feature for flexsurv.

[chjackson]

I agree that regression on cumulative incidences would be a good fit for the package, as this is a common approach to competing risks models. I'd implement it if I needed this class of models in my own work, which I don't at the moment but it's not out of the question that I will in the future!

I've also not reviewed what other software is available for these models - someone else may have done something related in R.