Bounding Omitted Variable Bias Using Auxiliary Data: With an Application to Estimate Neighborhood Effects

Abstract

This paper proposes a new estimator that bounds omitted variable bias using proxies
for omitted variables with an asymptotically valid bootstrap procedure. The proxies do
not need to appear in the same dataset as the outcome variable, and the estimator is
robust to measurement errors in proxies. I provide open-source software to implement
the estimator and its confidence interval. Next, I illustrate the application in the context of
estimating neighborhood effect on intergenerational cultural transmission and find that
growing up in an ethnic enclave could change adulthood outcomes for second-generation
immigrants that reflect their extent of cultural assimilation.

Keywords: Omitted Variable Bias, Two-Sample Least Squares, Data Combination, Data Fusion, Auxiliary Data, Proxy Variable, Intergenerational Cultural Transmission, Cultural Assimilation, Neighborhood Effect

JEL Classification: C01, C80, J01, J15, J16, J61, Z10, Z13