Common Errors: How to (and Not to) Control for Unobserved Heterogeneity

Review of Financial Studies, 2014, 27(2), 617-61

AFA 2013 San Diego Meetings Paper

58 Pages Posted: 17 Mar 2012 Last revised: 17 Jun 2014

Todd A. Gormley

Washington University in St. Louis

David A. Matsa

Northwestern University - Kellogg School of Management; National Bureau of Economic Research (NBER)

Date Written: August 3, 2013

Abstract

Controlling for unobserved heterogeneity (or “common errors”), such as industry-specific shocks, is a fundamental challenge in empirical research. This paper discusses the limitations of two approaches widely used in corporate finance and asset pricing research: demeaning the dependent variable with respect to the group (e.g., “industry-adjusting”) and adding the mean of the group’s dependent variable as a control. We show that these methods produce inconsistent estimates and can distort inference. In contrast, the fixed effects estimator is consistent and should be used instead. We also explain how to estimate the fixed effects model when traditional methods are computationally infeasible.

Additional programming advice can be found on our websites.

Keywords: unobserved heterogeneity, group fixed effects, industry-adjust, bias

JEL Classification: G12, G2, G3, C01, C13

Suggested Citation

Gormley, Todd A. and Matsa, David A., Common Errors: How to (and Not to) Control for Unobserved Heterogeneity (August 3, 2013). Review of Financial Studies, 2014, 27(2), 617-61; AFA 2013 San Diego Meetings Paper. Available at SSRN: https://ssrn.com/abstract=2023868 or http://dx.doi.org/10.2139/ssrn.2023868

Todd A. Gormley (Contact Author)

Washington University in St. Louis ( email )

One Brookings Drive
Campus Box 1133
St. Louis, MO 63130-4899
United States
(314) 935-7171 (Phone)

HOME PAGE: http://www.gormley.info

David A. Matsa

Northwestern University - Kellogg School of Management ( email )

2001 Sheridan Road
Evanston, IL 60208
United States
847-491-8337 (Phone)
847-491-5719 (Fax)

National Bureau of Economic Research (NBER) ( email )

1050 Massachusetts Avenue
Cambridge, MA 02138
United States

Paper statistics

Downloads
2,271
Rank
4,234
Abstract Views
10,484