Chris Jackson

`flexsurv::bootci.fmsm` can do multicore bootstrapping for any user-defined output from a flexsurv model, but there is no multicore feature in `standsurv` (copied to @mikesweeting as the author of this function).

This is a bit messy unfortunately! These functions have grown organically rather than being meticulously designed. `normbootfn.flexsurvreg` is not exposed to users, and is limited to bootstrapping the functions included...

That's a nice clear set of slides, thanks for finding. That all makes sense. That definition of the Cox-Snell residual assumes a proportional hazards model, but like the slides say,...

Go for it - I'm happy to have you contribute.

I'm having trouble following the theory behind this modification. Why would we favour adding any specific number to the residual for censored observations? What is meant by making them "compatible"?...

There isn't as far as I know, I'm afraid. In the meantime, I wonder if this commented code would be helpful to you or others. This is a three-state non-hidden...

For anyone interested in phase-type semi-Markov models, there is now an experimental package [msmbayes](https://chjackson.github.io/msmbayes/) which implements [these](https://chjackson.github.io/msmbayes/articles/advanced.html), and some of the other models in `msm`, using Bayesian inference in Stan....

Both of those sound good - thanks!

See [`help(msm)`](https://chjackson.github.io/msm/reference/msm.html), argument `"na.action"`. It uses all available data, basically. The definition of this is subtle though. If the last covariate value for a subject is `NA`, this record isn't...

There's no technique in flexsurv to do random effects that I know of. There is other software that would support this class of model e.g. [frailtypack](https://cran.r-project.org/web/packages/frailtypack/index.html) in R, or Bayesian...