fannie-mae

ETL process which downloads, transforms, and loads Freddie Mac/Fannie Mae mortgage data

Machine Learning project to assess the probability of default for Single Family homes in the United States

Fannie Mae loan-level data to ClickHouse

Integration with Fannie Mae's housing finance data API to retrieve current loan and income limits for any US county, census tract, or address

Predicting mortgage default risk with logistic regression in R, built on ~420K Fannie Mae single-family loans and written up as a reproducible Quarto report.

End-to-end credit risk modeling system using Fannie Mae data, including training pipelines, persisted model artifacts, and a Streamlit-based loan scoring app.

Mortgage credit scorecard validated out-of-time through the 2008 crisis. 19.1M loans, SAS.

Integration with Fannie Mae's APIs providing formulas for income limits and loan limits data by FIPS code or address

MCP server for Cal — the AI mortgage platform from HomeLoanExpress. 166 wholesale lenders, 1,400+ indexed program docs, 250 DPA programs. The first mortgage MCP server with a real broker-accessible lender data moat.

Federal Housing Finance Agency (FHFA) — independent third-party profile of a public API surface, by API Evangelist. The Federal Housing Finance Agency (FHFA) is an independent federal regulator established in 2008 that supervises Fannie Mae, Freddie Mac, and the Federal Home Loan Bank System. FHFA provides publicly accessible data APIs and datasets

Analysis using R and tidyverse to compare borrower behavior and characteristics between the years 2007 and 2019, focusing on key financial metrics such as credit scores, interest rates, debt to income ratios, and loan to value ratios.

IRB LGD/EAD framework on Fannie Mae mortgages: two-stage cure/severity models, downturn calibration and MoC per EBA guidelines.