Optimally hounded score functions for generalized linear models with applications to logistic regression
Leonard A. Stefanski, Raymond James Carroll, David Ruppert · Biometrika · 1986
We study optimally bounded score functions for estimating regression parameters in a generalized linear model. Our work extends results obtained by Krasker & Welsch (1982) for the linear model and provides a simple proof of Krasker & Welsch's first-order condition for strong optimality. The application of these results to logistic regression is studied in some detail with an example given comparing the bounded-influence estimator with maximum likelihood.
Cited by 132 works. doi:10.1093/biomet/73.2.413
Authors
References
- Generalized Linear Models
- Analysis of Binary Data
- The behavior of maximum likelihood estimates under nonstandard conditions
- The Influence Curve and Its Role in Robust Estimation
- Logistic Regression Diagnostics
- The Efficiency of Logistic Regression Compared to Normal Discriminant Analysis
- Efficient Bounded-Influence Regression Estimation
- Robust $M$-Estimators of Multivariate Location and Scatter
- Estimation in Linear Regression Models with Disparate Data Points
- One-Step Huber Estimates in the Linear Model