Online Program Home
  My Program

Abstract Details

Activity Number: 119 - SPEED: Bayesian Methods Student Awards
Type: Contributed
Date/Time: Monday, July 31, 2017 : 8:30 AM to 10:20 AM
Sponsor: Mental Health Statistics Section
Abstract #323662 View Presentation
Title: Stagewise Generalized Estimating Equations with Applications to Suicide Prevention
Author(s): Gregory Vaughan* and Robert Aseltine and Kun Chen and Jun Yan
Companies: Department of Statistics, University of Connecticut and Division of Behavioral Science and Community Health, UConn Health Center and Department of Statistics, University of Connecticut and University of Connecticut
Keywords: bi-level selection ; group selection ; penalized regression ; sparsity ; stagewise estimation
Abstract:

Stagewise estimation is a slow-brewing approach for model building that is very attractive in dealing with complex data structures for both its computational efficiency and its connection with penalized estimation. Using generalized estimating equations, we study stagewise estimation approaches that can handle clustered data and non-Gaussian models in the presence of a variable grouping structure. As important groups may still contain irrelevant variables, the key is to simultaneously conduct group selection and within-group variable selection. We propose two approaches to address the challenge. The first is a bi-level stagewise estimating equations (BiSEE) approach, which is shown to correspond to the sparse group lasso penalized regression. The second is a hierarchical stagewise estimating equations (HiSEE) approach, in which each stagewise estimation step itself is executed as a hierarchical selection process. Simulation studies show that BiSEE and HiSEE are competitive. We apply the proposed approaches to study the association between the suicide-related hospitalization rates of the 15--19 age group and the characteristics of the school districts in the State of Connecticut.


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

Back to the full JSM 2017 program

 
 
Copyright © American Statistical Association