Wei-Yin Loh
University of Wisconsin, Madison
Regression Tree Models for Multi-response and Longitudinal Data
Two extensions to the GUIDE classification and regression tree
algorithm will be presented. The first is prediction and modeling of
multi-response, longitudinal, or functional data and the second is
variable selection for identification of population subgroups for
differential treatment effects in personalized medicine. The solutions
can cope with large numbers of predictors and with missing observations.