Abstract
Wilks' test for a single outlier in a multivariate normal sample is extended to the case of samples from different subpopulations with common covariance matrix, a situation arising in MANOVA, for example. Simulation results show that the size of the test is acceptably robust to moderate heterogeneity in covariances (25–50% difference in total variation), especially if sample sizes are small (below 20 per group). However covariance heterogeneity leads to a drastic loss of power, unless this heterogeneity is concentrated in one dimension and the outlier appears in a different dimension. It is concluded that the extended test should be used with caution since it will often be difficult to establish whether these conditions hold.
Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callMore From: Communications in Statistics - Simulation and Computation
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
