November 16, 2017

Big Data Is a Big Deal at Freddie Mac

Michael Bradley
Michael Bradley, SVP of Single-Family Models and Analytics

Big data is the buzzword du jour. There’s no shortage of data in the mortgage finance industry but simply gathering a lot of data isn't useful. It only adds value if you can draw meaningful insights from it. Freddie Mac is harnessing the power of big data and Reimagining the Mortgage ExperienceSM to drive significant value for our customers, their borrowers and the industry.

Consider this:

  • There are 249 unique data points in the Uniform Closing Dataset (UCD), the latest common collection of data that accompanies each residential mortgage loan we buy. We collect the UCD in addition to the Uniform Appraisal Dataset (UAD) and Uniform Loan Delivery Dataset (ULDD).
  • Loan Advisor Suite® is our innovative, integrated end-to-end technology solution designed to provide lenders with greater certainty, usability, reliability and efficiency. Loan Advisor Suite assesses more than 300,000 unique single-family loan files and more than 712,000 appraisals monthly.
  • We have 40+ years of historical data, which amounts to more than 65 million mortgage records.

Freddie Mac is harnessing the power of big data and Reimagining the Mortgage ExperienceSM to drive significant value for our customers, their borrowers and the industry.

Big data typically refers to things that can be done on a large scale that cannot be done at a smaller scale, and Freddie Mac’s extremely large data sets can reveal patterns and trends when properly analyzed using advanced analytics. That’s why we’ve assembled a team of experts to make the most of it.

It starts with the data governance team, who confirms that the data we take in is complete and accurate. Data engineers make sense of messy data and make sure that information is in easy-to-use formats. Data scientists use machine learning to design ways to look for connections between pieces of information. And credit and risk experts leveraging algorithms and other technologies analyze this information to inform the development of products, tools and policies.

The launch of Loan Advisor Suite in July 2016 is a significant example of how we’re doing that, putting Freddie Mac’s risk assessment tools in our customers’ hands to simplify the process and reduce costs for lenders and, ultimately, for borrowers. Loan Advisor Suite allows lenders to apply Freddie Mac’s view of risk at any point in the origination process.

Since the launch, we’ve continued to use big data and advanced analytics to create powerful new methodologies and enhancements in Loan Advisor Suite to better meet the industry’s needs. These enhancements include introducing automated assessments of borrowers without credit scores, immediate collateral rep and warranty relief, and our automated collateral evaluation (ACE).

ACE, launched in June 2017, is unique to Loan Product Advisor<®, Loan Advisor Suite’s foundational tool. It uses proprietary algorithms that leverage a vast array of historical and public data to assess whether we can eliminate the need for a traditional property appraisal, which can save a borrower about $500 and decrease closing times by seven days or more.

In addition to helping consumers and lenders save time and money, we are using big data and machine learning to help us review all the appraisal data we receive from lenders through the Uniform Collateral Data Portal (UCDP). Using proprietary image processing algorithms, our team recently identified instances where the exact same interior photos were submitted to the UCDP for different sets of properties. This could indicate fraudulent activity; it could mean that the appraiser did not actually visit the property or that the appraiser made a mistake when uploading the photos. This use of machine learning enables us to identify these types of trends and patterns in a way and at a scale that human reviews cannot.

These are just a few of the ways our team of experts is using big data to simplify complex processes and deliver value to our customers and, in the process, build a better housing finance system for borrowers nationwide. And that’s a big deal.

  • Feedback

    Have a comment or question about this post? Email us to let us know what's on your mind.

    Maximum of 250 characters.