Mortgage Analysis

Data

This project uses Consumer Financial Protection Bureau (CPFB) data that is aggregated by state, metro and non-metro areas, and county. Dataset from CPFB which include the full time series and all revisions. These data cover January 2008–December 2021.

Data Source

These data come from the National Mortgage Database (NMDB), which uses a nationally representative sample of residential mortgages.

Mortgages 30–89 days delinquent

The 30-89 mortgage delinquency rate is a measure of early stage delinquencies and can be an early indicator of the mortgage market's overall health. It captures borrowers that have missed one or two payments.

Mortgages 90 or more days delinquent

The 90–day delinquency rate is a measure of serious delinquencies. It captures borrowers that have missed three or more payments. This rate measures more severe economic distress.

Technologies

• Python (Anaconda3) with Vistual Studio as preferred editor
• JupyterLab for analysis of data to discover trends for mortgages that are deliquent for metro and non-metro regions in the United States of America.
• Interactive StreamLit web application that can accept user input in the future.

How To

/mortgage_analysis.py

This file contains the functions to read the CSV data and return dataframes grouped by various columns.

/mortgage_stats.ipynb

This is a Jupyter Notebook with tables and visualizations of the mortgage data.

To run notebook

jupyter lab mortgage_stats.ipynb

/mortgage_analysis_webapp.py

This is a Streamlit page using the same data from the Jupyter Notebook

To run webapp

streamlit run mortgage_analysis_webapp.py