A note on testing AR and CAR for event studies

Return event studies generally involve several companies but there are also cases when only one company is involved. This makes the relevant testing problems, abnormal return (AR) and cumulative abnormal return (CAR), more difficult since one cannot exploit the multitude of companies (by using a relevant central limit theorem, say). We propose a permutation test that is valid under weaker conditions than the tests that have previously proposed in the literature in this context. We address the question of the power of the test via a brief simulation study and also illustrate the method with two applications to real data.