Single Firm / Single Event-Analyses: Robust Inference in Litigation

Event studies have become pivotal in securities fraud litigation. The typical use-case are single firm/single event (SFSE) applications. These differ in methodology from standard event studies in that inference on abnormal returns can only be based on the time-series variance of abnormal returns over a calibration period. In this study, we analyze robust inference in the SFSE setting using Monte Carlo and resampling experiments. We test size and power of different methods in different regimes of market volatility. It turns out that even extreme regimes of heteroscedasticity do not bias inference significantly when using daily data. The one problem of inference arises from switching volatility regimes, where calibration and event period occur in different regimes. We show that White- or Newey/West-correction of standard errors can not be applied and that GARCH estimation using intraday data is a suitable approach to solve this issue.