GLM
GLM copied to clipboard
bwlewis
R package notes
We'll roll all these methods in to an R package.
But, the SVD Newton method will be the primary non-distributed method. We should add the following to it:
- R-like subset selection
- Full set of output stats
Also note that the termination criterion is not the same as R, we should change that.
I'd add to the documentation at http://bwlewis.github.io/GLM/ more intuition about how does applying the G function helps. Namely if I do linear regression what would be the difference between minimizing the regular cost function to the logistic cost function.
