Hi,
Cross validation is a useful way to assess model performance in the real world. Basically you train the model with part of the data and then test with data that has been withheld so you get an estimate of accuracy.
http://en.wikipedia.org/wiki/Cross-validation_(statistics)
I was wondering if there was any native Cross Validation functionality in Hana PAL (reading the documents it looks like there isn't). Or if anyone has written their own procedures to do this, possibly in R (e.g. using caret)? It's something I'd really like to have and would write myself but I don't want to go reinventing the wheel
Thanks in advance,
Ian