Difference-In-Difference Design With Repeated Cross-Sections Under Compositional Changes: a Monte-Carlo Evaluation of Alternative Approaches
Type: preprint
Abstract: We discuss the potentially severe bias in the commonly-used Difference-in-Difference estimators under compositional changes and propose a Double Inverse Probability Weighting estimator for repeated cross-sections based on both the probability of being treated and of belonging to the post-treatment period, deriving also its doubly-robust version. Through Monte Carlo simulations, we compare its performance with several methods suggested by the literature. Results show that our proposed estimator outperforms all alternatives in the most realistic scenarios. We provide an empirical application estimating the effect of tariff reduction on bribing behavior on trades data between South Africa and Mozambique during the period 2006–2014.
Authors: Tommaso Manfè, Luca Nunziata
Language: English
Subfield: Statistics and Probability
Field: Mathematics
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W7111961541
