Inference with Difference-in-Differences Revisited

Arbeitspapier

A growing literature on inference in difference-in-differences (DiD) designs with grouped errors has been pessimistic about obtaining hypothesis tests of the correct size, particularly with few groups. We provide Monte Carlo evidence for three points: (i) it is possible to obtain tests of the correct size even with few groups, and in many settings very straightforward methods will achieve this; (ii) the main problem in DiD designs with grouped errors is instead low power to detect real effects; and (iii) feasible GLS estimation combined with robust inference can increase power considerably whilst maintaining correct test size – again, even with few groups.

Object Details

• Language: Englisch

References and Relationships

• Bibliographic citation: Series: IZA Discussion Papers ; No. 7742

Classification and Topics

• Classification: Wirtschaft ; Hypothesis Testing: General ; Estimation: General ; Single Equation Models; Single Variables: Cross-Sectional Models; Spatial Models; Treatment Effect Models; Quantile Regressions
• Subject: difference in differences ; hypothesis test ; power ; cluster robust ; feasible GLS

Contributors, Places and Time

• Event: Geistige Schöpfung
• (who): Brewer, Mike ; Crossley, Thomas F. ; Joyce, Robert
• Event: Veröffentlichung
• (who): Institute for the Study of Labor (IZA)
• (where): Bonn
• (when): 2013

Further information

• Handle: http://hdl.handle.net/10419/89904
• Last update: 10.03.2025, 11:44 AM CET