New Study Addresses Parameter Recovery in Severely Imbalanced Logistic Regression

A new arXiv paper examines how binary logistic regression parameter estimation and classification degrade when minority class observations are few, proposing zero-complement extrapolation methods for rank-deficient regimes.

Read the full story at arXiv stat.ML ↗

Timeline · 1 report

1. 2026-10-30 04:00 · arXiv stat.ML
   Zero-Complement Extrapolation: Parameter Recovery and Classification in Severely Imbalanced Logistic Regression under Rank-Deficient Regimes