Terry Therneau

This is serious stuff. What is surprising is that I can make it happen with R sometimes, but not when I turn on valgrind, the tool that would help me...

1. You gave a great example 2. I think time weights (Uno paper) is a tempest in a teacup so don't pay much attention to it, and thus never created...

This is a combination I hadn't seen or tested before: prediction + frailty + frailty has a missing value. I've found the issue, now have to figure out a solution....

Please take note of the very first lines in coxph.r. --- the file is automatically derived from the actual source in the now directory. Any patches have to be there....

Typo in the last message: the "noweb" directory. (Actually, I typed correctly but a spell checker intervened).