Ideas for Data Binning

Post date: Dec 17, 2013 8:16:53 AM

Binning data and using weight of evidence to transform the data into probability space is proven a very helpful process. However, there is no package in R or Python for doing so, yet SAS has the package.

I'm thinking about making my own data transformation package in R and Python using:

1. Simply binning and grouping
2. Using decision tree to partition each individual variable

Here are some interesting resources:

• Bin-summarise-smooth: A framework for visualising large data
• stats.bin: Bin function in R