Chris Jackson

@tdhock Is this the right link for the reviewer guide: (https://devguide.ropensci.org/reviewerguide.html). The link in the bot post gives a 404.

## Package Review - **Briefly describe any working relationship you have (had) with the package authors.** - None - ☒ As the reviewer I confirm that there are no [conflicts...

Yes in flexsurv, `predict(...,type="linear")` returns the "fitted values of the location parameter" - understood as being on the natural scale, not logged. Perhaps I should disambiguate this doc.

Both `flexsurvspline` and `flexsurvreg` return objects of class `"flexsurvreg"`, so the same predict method will be used. Is there a confusion here since flexsurv models can be based on different...

FYI if this is helpful, for a parameter named `"parname"` in a fitted flexsurv model `x`, the function `x$dlist$transforms[["parname"]]()` transforms a parameter value from its natural scale to the unrestricted...

No it's `x$dlist$transforms` to go from the natural scale to the unrestricted scale, and `x$dlist$inv.transforms` to go the other way.

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...

Thanks for the report. The default `newdata` that the flexsurvreg predict method uses is the "model frame" that is created in [this line of flexsurvreg.R](https://github.com/chjackson/flexsurv-dev/blob/master/R/flexsurvreg.R#L837). When run with a `ns()`...

This is proving tricky to handle. The function `stats::get_all_vars` seems like it would be useful here, as it is designed to extract the original variables supplied to a formula, whereas...

Not an official way, no. You could delete the $data component of the fitted model object, but various model output functions would then stop working. It's stored in order to...