A research on how macroeconomic, microeconomic factors and personal data could affect mortgage risk using Machine Learning techniques.
-
Updated
Nov 16, 2019 - Jupyter Notebook
A research on how macroeconomic, microeconomic factors and personal data could affect mortgage risk using Machine Learning techniques.
The goal of this project is to perform default prediction for commercial real estate property loans based on 17 variables.
A program to take in loan level data and create a model which can predict probability of default
PropIntel is an open-source, AI-driven property intelligence platform that tracks U.S. property ownership, notices of default, liens, foreclosures, auctions, tax delinquencies, and title transfers. It connects public records to reconstruct property histories, identify ownership patterns, and forecast real estate trends.
To associate your repository with the mortgage-defaults topic, visit your repo's landing page and select "manage topics."