The Home Mortgage Disclosure Act (HMDA) was enacted by Congress in 1975 and requires financial institutions like banks, savings associations, credit unions, and other mortgage lending institutions to report information about all of the loan applications they receive. In 2011, the authority for this reporting was transferred to the Consumer Financial Protection Bureau (CFPB).
This type of public disclosure data is important because it helps show whether lenders are serving the housing needs of their communities, gives public officials information that helps them make decisions and policies, and sheds light on lending patterns that could be discriminatory.
Mortgage approvals are intertwined with many other social concerns including employment, loan requirements, and income. Data science provides tools for predicting whether a mortgage application will be accepted, as well as for gaining insight into the factors that influence that decision. Digging into underlying patterns in HMDA data is the type of citizen science that can help provide greater visibility.
In previous capstone challenges, we looked at how poverty, income, health, ethnicity and other sociodemographic factors are related to evictions. Now, we are challenging you to consider how demographics, location, property type, lender, and other factors are related to the mortgage rate offered to applicants and to use your skills to build a model for predicting this rate for loan applications across the United States.