Nonparametric Multivariate Inference in High Dimensions - Xiaoli Kong
This event is in the past.
Loyola University Chicago
Nonparametric Multivariate Inference in High Dimensions
In this talk, a fully nonparametric (rank-based) method is introduced for comparing multiple groups. No assumption has been made on the distribution. We only require that the dependencies between the variables satisfy some mild conditions. In particular, to develop the theory, we prove a novel result for studying the asymptotic behavior of quadratic forms in ranks. The simulation results show that the developed rank-based method performs comparably well with mean-based methods. It has significantly superior power for heavy-tailed distribution with the possibility of outliers. The results are applied to Electroencephalograph (EEG) data that arose from a study to examine the correlates of genetic predisposition to alcoholism.