A bootstrap test for equality of variances

We introduce a bootstrap procedure to test the hypothesis Ho that K+1 variances are homogeneous. The procedure uses a variance-based statistic, and is derived from a normal-theory test for equality of variances. The test equivalently expressed the hypothesis as Ho:η=(η1,…,ηK+1)T=0, where ηi's are log contrasts of the population variances. A box-type acceptance region is constructed to test the hypothesis Ho. Simulation results indicated that our method is generally superior to the Shoemaker and Levene tests, and the bootstrapped version of the Levene test in controlling the Type I and Type II errors.