Conducting event studies with thinly traded stocks

Abstract

Procedures for measuring abnormal performance around events when securities do not trade daily are examined. A rank test patterned after the statistic proposed by Corrado [Corrado, C.J. (1989), A nonparametne test for abnormal security-price performance in event studies Journal of Financial Economics 23, 385-395] is derived. The empirical frequency distributions of the rank test statistic and a conventional test statistic are assessed under different rules for handling missing stock returns. The results suggest that the traditional procedures are reasonably well-specified for thickly and moderately traded stocks but misspecified for thinly traded stocks. The rank test however performs well for all trading frequencies and trade-to-trade returns are the best way to handle missing returns.