The study aims to identify an effective algorithm for selecting significant and confounding variables in logistic regression.
Development of a macro algorithm for variable selection in logistic regression.
Evaluation of the algorithm's performance against traditional methods.
Implementation in a controlled environment to assess retention of significant covariates.
The algorithm successfully retains significant covariates more effectively than standard approaches.
High retention rates of confounding variables were observed, enhancing model accuracy.
Statistical analysis indicates improved predictive performance with the new tool.
Abstract
If an analyst is in need of an algorithm that will help guide the retention of significant covariates as well as confounding ones they should consider this macro as an alternative tool.