Trusting Difference-in-Differences Estimates More: An Approximate Permutation Test
Year: 2017
Type: preprint
Abstract: In economics and business, policy researchers often use observational data and difference-in-differences estimation to test the effectiveness of a policy change. In this context, the model specification and the estimator for standard errors are crucial for correct inferences, given unobserved heterogeneity and correlation across firms and time. Existing literature mainly focuses on the correct size of inferential tests. We suggest a non-parametric test based on the empirical error distribution of the estimated law effect, which we reveal with placebo laws. In comparison to alternative parametric tests, our procedure improves power, under the stringent condition of maintaining a correct size.
Source: SSRN Electronic Journal
Authors: Sebastian Bunnenberg, Steffen Meyer
Institutions: Reutlingen University, University of Southern Denmark
Language: English
Subfield: Statistics and Probability
Field: Mathematics
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W2587235528
