|
Activity Number:
|
345
- High-Dimensional Statistics
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Tuesday, July 30, 2019 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
IMS
|
|
Abstract #306611
|
|
|
Title:
|
Divergence Based Inference for High-Dimensional GLMM
|
|
Author(s):
|
Lei Li* and Anand Vidyashankar
|
|
Companies:
|
George Mason University and George Mason University
|
|
Keywords:
|
GLMM;
Divergence Method;
Contingency Tables;
Random Effects;
Robust
|
|
Abstract:
|
High dimensional generalized linear mixed models (GLMM) arise in several contemporary applications spanning various scientific disciplines. Robust and Efficient inference in these problems are challenging due to heavy computational burden and theoretical challenges. Specifically, robustness analyses require investigation of the effect of choice of the random effect distribution on the parameter estimates. In this presentation we describe a new divergence based methodology which allows for a large class of distributions for the random effect distribution. Using techniques from convex analyses, we establish both theoretical and computational properties of our proposed procedure. We illustrate with an application to high-dimensional sparse contingency table analysis.
|
Authors who are presenting talks have a * after their name.
