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

By Todd A. Gormley, David A. Matsa. 2014

1,057 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/3fd17a19831e373cc21214c22e33322c9796fb16

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Economics and Econometrics

Citations in the corpus

Corporate Finance Policies and Social Networks (2017, methodology): “We control for industry leverage by adding industry-year dummies, instead of controlling for industry leverage to control for unobserved heterogeneity, following Gormley and Matsa (2014).”
Do Debt Contract Enforcement Costs Affect Financing and Asset Structure (2014, positive): “In particular, we repeat our tests after including industry-year fixed effects (with industry defined at the two-digit NIC code level), following Gormley and Matsa (2014), and find our results to be robust (Column 6).”
Customer Concentration Risk and the Cost of Equity Capital (2016, background): “Columns 1-3 report the results from regression models that include all control 12 For a full discussion of how fixed effects can remove an omitted variable bias, see Gormley and Matsa (2014).”
Social Capital, Trust, and Firm Performance: The Value of Corporate Social Responsibility during the Financial Crisis: Social Capital, Trust, and Firm Performance (2017, positive): “22 Gormley and Matsa (2014) recommend the inclusion of further fixed effects to control for unobserved firm heterogeneity.”
An Efficiency Perspective on Carbon Emissions and Financial Performance (2020, methodology): “…by Λ , which includes year, industry, and country-fixed effects, to rule out potential confounding from unobserved factors that might drive both carbon efficiency and financial performance over time and across sectors and countries (Fama and French, 1997; Gormley and Matsa, 2014; Horváthová, 2010).”
Estimating Dynamic Panel Models in Corporate Finance (2013, methodology): “11 Gormley and Matsa (2012) show that the inclusion of the industry median, a common practice in capital structure research, introduces measurement error bias.”