Bootstrap Tests for Multivariate Event Studies

Statistical tests for multivariate event studies, exact or asymptotic, have been derived based on multivariate normality. As it has been documented that the performances of these tests are not satisfactory because stock returns are far from normally distributed, especially for daily returns, this paper proposes a bootstrap alternative to multivariate event studies that does not require a specific distributional assumption. The Monte Carlo experiments based on daily returns data show that the bootstrap tests outperform the traditional tests by having close rejection rates to the nominal significance levels. The traditional tests, in contrast, tend to reject the null hypotheses too often