Activity Number:
|
464
- SPEED: Infectious Diseases, Spatial Modeling and Environmental Exposures, Speed 1
|
Type:
|
Contributed
|
Date/Time:
|
Wednesday, July 31, 2019 : 8:30 AM to 10:20 AM
|
Sponsor:
|
Section on Statistics in Epidemiology
|
Abstract #305092
|
|
Title:
|
A Method for High-Dimensional Variable Selection in Presence of Collinearity
|
Author(s):
|
Jiyeong Jang* and Sanjib Basu
|
Companies:
|
University of Illinois at Chicago and University of Illinois at Chicago
|
Keywords:
|
Collinearity;
Environmental exposure;
Variable selection
|
Abstract:
|
Variable selection has become an essential element of statistical modeling to yield parsimonious models while keeping high prediction accuracy. Particularly, as the size of the data increases tremendously, variable selection can be challenging in presence of collinearity among features. For example, in the setting of understanding the role of environmental exposures on health outcomes, the number of environmental exposure variables are often large and are strongly inter-correlated. We develop a method for variable selection in high dimensional data which clusters strongly correlated variables to efficiently handle the collinearity problem. The performance of this method is evaluated with respect to multiple performance criteria in an extensive set of simulation studies. We apply this method to analyze the role of environmental exposures in real-world data.
|
Authors who are presenting talks have a * after their name.