Influence-Weighted Adaptive LASSO for Robust Variable Selection in Logistic Regression: Research Code
This software archive contains the Python implementation associated with the research study entitled "Influence-Weighted Adaptive LASSO for Robust Variable Selection in Logistic Regression: A Simulation Study with Correlated Predictors and Data Contamination." The code implements an influence-weighted adaptive LASSO procedure for logistic regression. It includes data preprocessing, cross-fitted residual diagnostics, robust leverage estimation, observation weighting, adaptive penalty construction, model fitting, cross-validation, and contamination experiments. The experiments consider vertical contamination, bad-leverage contamination, and good-leverage contamination using the Wisconsin Breast Cancer Diagnostic dataset. The archive is intended to support reproducibility and facilitate further research on robust variable selection in logistic regression.
Authors
- Vimala Devi
Publication Details
- Journal
- Zenodo (CERN European Organization for Nuclear Research)
- Published
- 2026-11-20
- DOI
- https://doi.org/10.5281/zenodo.24393599
- Primary Topic
- Statistical Methods and Inference
- Type
- article
- Field-Weighted Citation Impact
- 0.00