Online Program Home
  My Program

All Times EDT

Abstract Details

Activity Number: 485 - A Unified View on Model Selection
Type: Contributed
Date/Time: Thursday, August 6, 2020 : 10:00 AM to 2:00 PM
Sponsor: Biometrics Section
Abstract #313030
Title: Spike-And-Slab LASSO for Estimation and Variable Selection in Longitudinal Mixed Endpoints Data with High-Dimensional Covariates
Author(s): Yunusa Olufadi* and Ebenezer Olusegun George
Companies: University of Memphis and University of Memphis
Keywords: Variable selection; Spike-and-slab ; Mixed endpoints; high-dimensional regression; Joint model estimation

Multiple endpoints of different kinds (discrete, continuous, count, ordinal) are ubiquitous in biomedical and biopharmaceutical studies. This is usually because a single endpoint is not adequate to describe the disease complexities and associations among these multiple endpoints are important in characterizing the treatment effectiveness. There are usually several levels of correlations inherent in such mixed outcomes, especially when dealing with a clustered or longitudinal design and procedures which analyze each outcome separately ignore these correlations and present misleading results. In this presentation, we introduce spike-and-slab LASSO with mixed endpoints (SSLME) to guide the estimation and efficient extraction of active predictors in longitudinal studies with high-dimensional covariates. We illustrate our methodology through extensive simulations analysis of real data.

Authors who are presenting talks have a * after their name.

Back to the full JSM 2020 program