Colloquium: Broken Adaptive Ridge Method for Variable Selection in Generalized Partly Linear Models with Application to the Coronary Artery Disease Data

University of Saskatchewan Department of Mathematics & Statistics Colloquium

The Colloquium Committee cordially invites you to attend a pre-term talk in the 2024-25 Math & Stats Colloquium series which will occur as a hybrid talk with the possibility to attend either in-person or in Zoom. We encourage everyone to consider attending in person if possible to help our speaker feel welcome and appreciated.

Date/Time: THURSDAY, August 22 @ 3:00 PM CST
(please make note of the non-standard day and time!)

Location: ARTS 206
Zoom link: Please contact colloquium@math.usask.ca

Speaker: Xuewen Lu, University of Calgary

Title: Broken Adaptive Ridge Method for Variable Selection in Generalized Partly Linear Models with Application to the Coronary Artery Disease Data

Abstract: Motivated by the CATHGEN data, we develop a new statistical learning method for simultaneous variable selection and parameter estimation under the context of generalized partly linear models for data with high-dimensional covariates. The method is referred to as the broken adaptive ridge (BAR) estimator, which is an approximation of the L0-penalized regression by iteratively performing reweighted squared L2-penalized regression. The generalized partly linear model extends the generalized linear model by including a non-parametric component to construct a flexible model for modeling various types of covariate effects. We employ the Bernstein polynomials as the sieve space to approximate the non-parametric functions so that our method can be implemented easily using the existing R packages. Extensive simulation studies suggest that the proposed method performs better than other commonly used penalty-based variable selection methods. We apply the method to the CATHGEN data with a binary response from a coronary artery disease study, which motivated our research, and obtained new findings in both high-dimensional genetic and low-dimensional non-genetic covariates.

Speaker Bio: Dr. Xuewen Lu is a Professor in the Department of Mathematics and Statistics at the University of Calgary. He earned his Ph.D. in Statistics from the University of Guelph in 1997. Prior to joining the University of Calgary, he served as a biostatistician at Agriculture and Agri-Food Canada for four years. Dr. Lu has held the position of Associate Head of Research in the Department of Mathematics and Statistics at the University of Calgary. He currently serves as the Regional Representative of the International Chinese Statistical Association (ICSA) - Canada Chapter for Canada West.

To date, Dr. Lu has supervised the research of 13 Ph.D. students, 27 M.Sc. students, and 3 postdoctoral fellows. In 2015, he was honoured with the FGS Great Supervisor Award from the University of Calgary. Dr. Lu is an active researcher, dedicated to the development of innovative statistical methodologies for survival analysis, longitudinal data analysis, mixed models, empirical likelihood methods, quantitative risk assessment, predictive microbiology models, and more. He has published over 130 papers in peer-reviewed journals and has served on the editorial boards of several esteemed statistical journals.

Please note that, because of the non-standard day and time, attendance at this talk is encouraged but not compulsory for graduate students.

Please contact colloquium@math.usask.ca if you have any questions. We hope to see you all on Thursday the 22^nd at 3:00 PM in ARTS 206!

Colloquium Committee
University of Saskatchewan
Department of Mathematics and Statistics
Find us on Facebook @uofs.math.stats and Twitter @UofS_Math_Stats