Improved Methods for Tests of Long-Run Abnormal Stock Returns

Abstract

Barber and Lyon (1997a) and Kothari and Warner (1997) document that standard tests of long-run abnormal returns are misspecified. In this research, we evaluate alternative methods to test for long-run abnormal returns. We document that two general approaches yield well-specified test statistics in random samples. The first approach uses a traditional event study framework and buy-and-hold abnormal returns calculated using carefully constructed reference portfolios, such that the population mean abnormal return is identically zero. Inference is based on either a skewness-adjusted t statistic or the empirically generated distribution of mean long-run abnormal returns. The second approach is based on the calculation of mean monthly abnormal returns using calendar-time portfolios and a time-series t statistic. Though both approaches perform well in random samples, misspecification in nonrandom samples is pervasive. Our central message is that the analysis of long-run abnormal returns is treacherous.