|
Activity Number:
|
608
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Wednesday, August 3, 2016 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Section on Statistical Learning and Data Science
|
|
Abstract #321367
|
View Presentation
|
|
Title:
|
Sparsity-Oriented Importance Learning
|
|
Author(s):
|
Chenglong Ye* and Yi Yang and Yuhong Yang
|
|
Companies:
|
University of Minnesota and McGill University and University of Minnesota
|
|
Keywords:
|
variable importance ;
Model mixing ;
Model selection
|
|
Abstract:
|
Model importance of variables in a dataset has attracted much attention in recent years. This paper proposes a new variable importance measure( SOIL) to address the issue, by the method of model mixing . Then we demonstrate the propeties of SOIL importance in contrast to other importance measures using several simulation models. Also we study two real data examples Colon cancer data and BGSboys date to provide some new insight of the SOIL importance.
|
Authors who are presenting talks have a * after their name.
