April 6, 2023 Webinar
Partially-global Frechet regression and its variable selection
Yichao Wu, PhD
Abstract
We propose a partially-global Frechet regression model by extending the profiling technique for the partially linear regression model (Severini and Wong 1992). This extension allows for the response to come from a generic metric space and can further incorporate predictors from another generic metric space, scalar predictors, or combinations of the two. By melding together the local and global Frechet regression models proposed by Petersen and Muller (2019), we gain a model that is more flexible than global Frechet regression and more accurate than local Frechet regression when the data generating process is truly ''global (linear)" for some scalar predictors or relies on non-Euclidean predictors. If time permits, I will also briefly talk about variable selection for partially-global Frechet regression.
Short Bio
Dr. Yichao Wu received his Ph.D. in statistics from the University of North Carolina at Chapel Hill in 2006. He joined the University of Illinois at Chicago in 2017 after serving many years at North Carolina State University. His research interest mainly lies in the general area of statistical machine learning and high dimensional data analysis.