Open JackStat opened 9 years ago
Patches welcome!
Revisiting this a little while later...
I'm not sure we've yet come to a really good way of handling distributions with both centrality and scale parameters that have to be fit (huberized, t are a bit of a mess). I could (in principle) see how to do the exponential distribution, but Weibull might be slightly more fun.
I think it is awesome that gbm supports survival analysis and I was wondering if it would be easy enough to add support for survival regression.