Applied Mathematics Colloquia by Arne Bathke: Nonparametric Inference Methods for Multivariate Data
Speaker:
, professor of artificial intelligence and human interfaces, University of Salzburg
Title:
Nonparametric Inference Methods for Multivariate Data
Abstract: When there are multiple response variables (endpoints) and different predictors, one typically wants to find out which predictors are relevant, and for which endpoints. We present two rather general approaches towards valid inference for multivariate data, one accommodating binary, ordinal, and metric endpoints, and the other allowing for a factorial design structure. The first approach is fully nonparametric, resulting in rank-based statistics and an F-approximation of the sampling distribution, while the second approach employs asymptotically valid resampling techniques (bootstrap). We also try to address the question of how well the proposed methods actually accomplish their goals, and how to use the respective toolboxes that have been developed.
Applied Mathematics Colloquium