Data source
Data from Freddie Mac, delivered clean.
4 datasets pulled from Freddie Mac's releases, checked field by field and shipped the way you want them — daily, weekly, or hourly, your call.
- 4 datasets
- 1 industry
- Real rows on request
What Datadory delivers from Freddie Mac
4Freddie Mac Single-Family Loan-Level Dataset
FHFA Public Use Database (PUDB)
FHFA House Price Index (HPI)
Ginnie Mae Disclosure Data and Reports
Pick a catch, see the rows.
Name any Freddie Mac dataset and we send real rows from it — not a screenshot of rows. 1,744 datasets. Pick your catch.
Get a sampleAPI, files, or your warehouse. Daily, weekly, or hourly.
Straight answers about Freddie Mac data
What is the Freddie Mac Single-Family Loan-Level Dataset?
Freddie Mac's flagship credit-performance disclosure: loan-level records on all single-family mortgages it purchased or guaranteed from January 1, 1999 through March 31, 2026 - about 56 million loans. Each quarter ships an origination file and a matching monthly performance file, split into a Standard book resembling CRT-eligible criteria and a Non-Standard book for everything else.
How far back does the loan performance history reach?
To January 1999. Every acquisition since then appears with monthly reporting through March 31, 2026 under Release 47 - twenty-seven years spanning the pre-2008 run-up, the crisis dispositions, HARP-era refinancings, Home Possible originations since 2015 and the post-pandemic rate cycle, all in one consistent layout.
What separates the Standard from the Non-Standard dataset?
Eligibility rules modeled on credit-risk-transfer criteria. The Standard book holds fully amortizing fixed-rate mortgages with full documentation, plus Relief Refinance and Home Possible loans funded on or after March 2015; government-insured FHA, VA and GRH loans sit outside both. Non-Standard carries typically ineligible loans, so models train only on the population they were designed for.
Can I test pipelines before handling the full tape?
That is what the sample dataset is for: 50,000 randomly selected loans per vintage year with fields identical to the full release. Code validated on it runs unchanged against the complete population, which is how most Datadory customers start - prototype small, scale once the joins are proven.