On The Use Of The Multivariate Regression-Model In Event Studies

A Multivariate Regression Model (MVRM) methodology to measure the effect of new information on asset prices was first suggested in Gibbons [1980, appendix H]. In this paper I outline the use of that methodology to measure abnormal returns and to test hypotheses about these returns. Included are advantages of the MVRM methodology over other event study methodologies and some of its problems in hypothesis testing.1 Section 2 discusses the MVRM technique relative to better-known methodologies. In section 3 I compare these methodologies and argue that the primary advantage of the MVRM is in testing joint hypotheses. Section 4 provides small sample evidence on several test statistics used in the MVRM; section 5 summarizes the paper.