Using Machine Learning to Estimate Racially Disaggregated Wealth Data at the Local Level

Understanding wealth is central for uncovering the barriers to wealth-building and designing policies that unlock opportunities for everyone. However, household wealth data at the local level are generally not widely available, especially statistics disaggregated by race and ethnicity. In this resea...

Full description

Saved in:
Bibliographic Details
Published in:Policy File 2023
Main Authors: Williams, Aaron R, Zhong, Mingli, Braga, Breno
Format: Report
Language:English
Subjects:
Intelligence
Urban Institute
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Understanding wealth is central for uncovering the barriers to wealth-building and designing policies that unlock opportunities for everyone. However, household wealth data at the local level are generally not widely available, especially statistics disaggregated by race and ethnicity. In this research report, we document how we use machine learning to estimate net worth and emergency savings data at the local, city, state, and national levels. We also disaggregate our estimates by racial and ethnic groups at the city, state, and national levels. Using a random forest model, we predict whether households in the American Community Survey have $2,000 in emergency savings and their net worth. We then aggregate this household-level data to produce statistics at different geographic levels and by racial and ethnic groups.