A Note on Local Likelihood Regression for Binary Response Data

Abstract

This article considers identifying the existence of the local likelihood estimator in binary regression. A simple method for the identification is proposed, which is derived by recognizing the problem as a linear complementarity problem through a support vector machine problem with soft margins.

Keywords:

• Binary regression
• Local likelihood estimator
• Support vector machine

Mathematics Subject Classification:

• Primary 62G08
• Secondary 90C90

Acknowledgment

The author thanks Kanta Naito and two reviewers for their helpful comments and suggestions.