Monotonic Binning Driven by Decision Tree
This article is originally published at https://statcompute.wordpress.com
After the development of MOB package (https://github.com/statcompute/MonotonicBinning), I was asked by a couple users about the possibility of using the decision tree to drive the monotonic binning. Although I am not aware of any R package implementing the decision tree with the monotonic constraint, I did manage to find a solution based upon the decision tree.
The Rborist package is an implementation of the Random Forest that would enforce the monotonicity at the local level within each tree but not at the global level for the whole forest. However, with a few tweaks on the Rborist syntax, it is not difficult to convert the forest with many trees into the forest with a single tree. After all necessary adjustments, I finally ended up with a decision tree that can be used to drive the monotonic binning algorithm, as shown in the arb_bin() function below, and will consider adding it into the MOB package later.
Please visit source website for post related comments.