|
Activity Number:
|
130
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Monday, August 1, 2016 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Section on Statistics in Epidemiology
|
|
Abstract #320158
|
View Presentation
|
|
Title:
|
High-Dimensional Analysis of Spatial Count Data: A Penalized Estimating Equation Approach
|
|
Author(s):
|
Rejaul Karim* and Tapabrata Maiti
|
|
Companies:
|
Michigan State University and Michigan State University
|
|
Keywords:
|
High-dimensional analysis ;
estimating equation ;
Poisson regression
|
|
Abstract:
|
Spatial count data arises in many scientific applications and Poisson regression is a commonly used approach for such data analysis.Under spatial dependence estimating equation is computationally simple for model parameter estimation. Traditional approach is suitable for small number of covariates and for relatively small data sets. In this talk, we extend the approach to high dimensional set up that are suitable to handle large data. Specifically we develop a penalized estimating equation method. Further we study both theoretical and numerical properties of the proposed approach and compare with other competitive methods.
|
Authors who are presenting talks have a * after their name.
