|
|
 |
| |
|
|
|
Bayesian Relative Importance Analysis of Logistic Regression Models |
|
|
|
PP: 53-69 |
|
|
|
doi:10.18576/jsapl/030201
|
|
|
|
Author(s) |
|
|
|
Xiaoyin Wang,
|
|
|
|
Abstract |
|
|
| The goal of determining the relative importance of predictors is to expose the individual contribution of the predictor in the presence of other predictors within a selected model. In practice, it is often desired to understand the extent to which each predictor variable drives the response variable The purpose of this article is to expand the current research practice to evaluate the relative importance of each predictor in a logistic regression setting by developing a statistical model-based approach in the Bayesian framework. Results from extended simulation studies suggest that the proposed weighted paired comparison model with the two sided power (TSP) link function provides the most effective and reliable measure of the relative importance of predictors. |
|
|
|
|
|
|
|
