Robust Inference in Single Firm / Single Event-Analyses
52 Pages Posted: 6 Feb 2019 Last revised: 21 Mar 2022
Date Written: February 8, 2022
Abstract
Single firm/single event (SFSE) applications of event studies matter in corporate finance, e.g.,due to their role in securities fraud litigation. Since inference on abnormal returns then has to rely on the time series variance of abnormal returns, the implied problem of heteroscedasticity is obvious but hard to solve. We analyze robust inference in the SFSE setting using Monte Carlo and resampling experiments. Estimation is biased when calibration and event period occur in different volatility regimes. We develop a unique specification test for such structural breaks. Most robust inference results from using intraday data and a multiplicative component GARCH estimator.
Keywords: Event studies, Inference, Monte Carlo simulation, Volatility, Structural breaks
JEL Classification: G01, G32, K41
Suggested Citation: Suggested Citation