On some test statistics for testing homogeneity of variances: a comparative study |
| |
Authors: | Dinesh Sharma B. M. Golam Kibria |
| |
Affiliation: | 1. Department of Mathematics and Statistics , James Madison University , 305 Roop Hall, MSC 1911, Harrisonburg , VA , 22801 , USA sharmadr@jmu.edu;3. Department of Mathematics and Statistics , Florida International University , Miami , FL , 33193 , USA |
| |
Abstract: | ![]() In this paper, we have reviewed 25 test procedures that are widely reported in the literature for testing the hypothesis of homogeneity of variances under various experimental conditions. Since a theoretical comparison was not possible, a simulation study has been conducted to compare the performance of the test statistics in terms of robustness and empirical power. Monte Carlo simulation was performed for various symmetric and skewed distributions, number of groups, sample size per group, degree of group size inequalities, and degree of variance heterogeneity. Using simulation results and based on the robustness and power of the tests, some promising test statistics are recommended for practitioners. |
| |
Keywords: | ANOVA homogeneity of variances Levene's test Monte Carlo simulation power of test Robustness Type I error rate |
|
|