Event studies with a contaminated estimation period

摘要: Event studies are an important tool for empirical research in Finance. Since the seminal contribution of Fama et al. [Fama, E., Fisher, L., Jensen, M., Roll, R., 1969. The adjustment stock prices to new information. International Economic Review 10, 1–21], there have been many enhancements classical test methodology. Somewhat surprisingly, estimation period has attracted less interest. It is usually routinely determined as a fixed window prior event announcement day. In this study, we propose that reduces impact potentially unrelated events during period. Our proposition based on two-state version market model return-generating process. We present standard specification and power analyses. results highlight importance explicitly controlling occurring window, especially presence event-induced increase return volatility.