Terry Therneau

One of the few things I did right in my long career was that "out of the office" = "no email". Neither read it or respond. Enjoy your time away.

I'm still trying to find out how to have a more general conversation. A bug list just doesn't seem correct. Comments a. There is no vignette here, nor any reference...

The issue of centering is one that resurfaces every dozen years in a new guise. As an example look at the longley data (type "longley" in R); James Longley, An...

Good catch. I hadn't thought about ymax/ymin when doing the 'trickery' code. In my defense, I can't think of a case where I would want to use ymin or ymax...

Why are you asking me about a failure in another package? Ask the author of emmeans.

I have very little sympathy for non-syntactic names, first of all. It's along the lines of my argument that "A_very_long_file_name_is_not_a_substitute_for_documentation". Second, and more importantly, I do all the formula processing...

1. This issue you raise is essentailly a documentation flaw. The low level routine match.ratetable checks all the user's input against the rate table for consistency: does the sex variable...

What do you want the prediction to be from a frailty model with new data? I'm not trying to be difficult -- I have struggled with this question myself. Say...

1. I think this is a sensible idea. 2. I would not include a t0 argument, since the survreg models do not allow for delayed entry. Someone can modify the...

The real issue is three fold: in this situation survfit can create a curve that is nonsense, doing otherwise would require a a level of artificial intelligence in the routine...