predict_foreclosure

Data cleaning/ ML project to predict foreclosures in Fannie Mae mortgages.

Data and Directory Structure

Once downloaded, the project expects to have all Python files in a "code" subdirectory and all data in a "data" subdirectory on the same level. The data can be downloaded here: http://www.fanniemae.com/portal/funding-the-market/data/loan-performance-data.html.

You will need all of the performance and acquisition data.

Requirements

Python version 2.4.14 Pandas version 0.23.1 Sklearn version 0.19.1

Running the reduction code

To run the reduction script reduce_data.py, first import it

import reduce_data

then run the script

reduce_data.main()

It will then produce a file named reduced_data.csv that will be a csv text file.