Estimating Family Income from Administrative Banking Data: A Machine Learning Approach
- (pp. 36-41)
AbstractThe JPMorgan Chase Institute uses administrative banking data for research. In order to address representativeness in our data, we seek a reliable estimate of gross family income for population segmenting and reweighting purposes. JPMC Institute Income Estimate (JPMC IIE) version 1.0 uses gradient boosting machines (GBM) to estimate gross family income based on a truth set drawn from credit card and mortgage application data. The estimation relies on administrative banking data in combination with zip code-level characteristics available through public datasets. The final model yielded a significantly more accurate prediction of income than checking account inflows alone.
CitationFarrell, Diana, Fiona Greig, and Erica Deadman. 2020. "Estimating Family Income from Administrative Banking Data: A Machine Learning Approach." AEA Papers and Proceedings, 110: 36-41. DOI: 10.1257/pandp.20201057
- G21 Banks; Depository Institutions; Micro Finance Institutions; Mortgages
- G51 Household Saving, Borrowing, Debt, and Wealth