A new robust approach for multinomial logistic regression with complex design model

TL;DR

This paper introduces a robust estimation and testing method for multinomial logistic regression models with complex design, using phi-divergence measures to enhance robustness and reliability.

Contribution

It develops new robust estimators and Wald-type tests based on phi-divergence measures specifically for multinomial logistic regression with complex sampling designs.

Findings

Robust estimators have bounded influence functions.

The proposed methods outperform traditional ones in simulations.

Numerical examples demonstrate practical effectiveness.

Abstract

Robust estimators and Wald-type tests are developed for the multinomial logistic regression based on $\phi$-divergence measures. The robustness of the proposed estimators and tests is proved through the study of their influence functions and it is also illustrated with two numerical examples and an extensive simulation study.