← Search

Inference for Heterogeneous Effects using Low-Rank Estimation of Factor Slopes

Victor Chernozhukov¹; Christian Hansen²; Yuan Liao³; Yinchu Zhu⁴

¹ Department of Economics, MIT, Cambridge, MA 02139. [email protected] · ² University of Chicago, Chicago, IL 60637. [email protected] · ³ Rutgers University. New Brunswick, NJ 08901, USA. [email protected] · ⁴ Brandeis University, Waltham, MA 02453. [email protected]

Abstract We study a panel data model with heterogeneous effects, allowing slopes to vary across individuals and time. To reduce dimensionality, we assume these slopes follow a factor structure, so slope matrices can be estimated via low-rank regularized regression. We propose a multi-step estimation procedure incorporating sample splitting and partialing-out to enable valid inference after penalized estimation. We establish the asymptotic normality of the resulting estimator, facilitating inference for individualtime- specific effects and their cross-sectional averages. The method’s performance is illustrated through simulations and an empirical application.

DOI

10.1162/rest.a.1735

Pages

1-45

Language

en

Export

BibTeX

Sources

openalex crossref