Professor Jialiang Li, from the Department of Statistics and Data Science and the Duke-NUS Graduate Medical School at the National University of Singapore, is a renowned researcher specializing in threshold models, structural equation modeling, personalized and diagnostic medicine, model averaging, data smoothing, statistical machine learning, and survival analysis. An elected member of the International Statistical Institute (ISI) and a fellow of the American Statistical Association (ASA) and the Institute of Mathematical Statistics (IMS), he has received accolades such as Singapore's Research Talent Award. Professor Li has published extensively in leading journals and serves as an editor for esteemed publications like the Annals of Applied Statistics, Biometrics, and the Annual Review of Statistics and Its Application.

Seminar Details:

• • Title: Robust Model Averaging Prediction of Longitudinal Response with Ultrahigh-dimensional Covariate
  • Speaker: Professor Dr. Jialiang Li
  • Time: 02:00 PM, Saturday, December 12^nd, 2024.
  • Venue: Room C.41, University of Science, VNU-HCM.
    (227 Nguyen Van Cu Street, Ward 4, District 5, Ho Chi Minh City)

Abstract:

Model averaging is an attractive ensemble technique to construct fast and accurate prediction. Despite of having been widely practiced in cross-sectional data analysis, its application to longitudinal data is rather limited so far. We consider model averaging for longitudinal response when the number of covariates is ultrahigh. To this end, we propose a novel two-stage procedure in which variable screening is first conducted and then followed by model averaging. In both stages, a robust rank-based estimation function is introduced to cope with potential outliers and heavy-tailed error distributions, while the longitudinal correlation is modeled by a modified Cholesky decomposition method and properly incorporated to achieve efficiency. Asymptotic properties of our proposed methods are rigorously established, including screening consistency and convergence of the model averaging predictor, with uncertainties in the screening step and selected model set both taken into account. Extensive simulation studies demonstrate that our method outperforms existing competitors, resulting in significant improvements in screening and prediction performance. Finally, we apply our proposed framework to analyze a human microbiome dataset, showing the capability of our procedure in resolving robust prediction using massive metabolites.

Keywords: Robust statistics; Longitudinal model; high-dimensional model

Registration:

We kindly invite faculty members, PhD candidates, graduate students, and undergraduate students to participate.
Register here!