Main Article Content
This study aims to evaluate the type I error rates and power of the nonparametric bootstrap test (tb0) for equality of covariance matrices of two dependent multivariate normal populations in order to compare its performance with the test presented by Jiang and Sarkar (1998) (W2 and W5) and Jiang et al. (1999) (LRT, LRT1, LRT2 and LRT3). For this simulations Monte Carlo were performed, considering the number of variables (p), sample sizes (n), covariance matrices (Σ) and signicance level (α) of 0.05. In the first case, for p = 2, it was concluded that among the tests that controlled the type I error, the tests tb0 , LRT3 and W2 were greater than its competitors in all cases studied. In relation to power, the test tb0 approached the testing LRT3 and W2 and is considered intermediate. In the second case, it is considered that p = 4 and p = 10, it was concluded that the test tb0 showed high performance, in most cases equal to 100% even for small samples (n = 20). Therefore, we recommend the application of the proposed test tb0 in real situations.
Authors who publish with this journal agree to the following terms:
- Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work (See The Effect of Open Access).