Improving the Efficiency of Robust Estimators for the Generalized Linear Model

The distance constrained maximum likelihood procedure (DCML) optimally combines a robust estimator with the maximum likelihood estimator with the purpose of improving its small sample efficiency while preserving a good robustness level. It has been published for the linear model and is now extended to the GLM. Monte Carlo experiments are used to explore the performance of this extension in the Poisson regression case. Several published robust candidates for the DCML are compared; the modified conditional maximum likelihood estimator starting with a very robust minimum density power divergence estimator is selected as the best candidate. It is shown empirically that the DCML remarkably improves its small sample efficiency without loss of robustness. An example using real hospital length of stay data fitted by the negative binomial regression model is discussed.

[1]  Alfio Marazzi,et al.  A robust conditional maximum likelihood estimator for generalized linear models with a dispersion parameter , 2018, TEST.

[2]  Victor J. Yohai,et al.  High finite-sample efficiency and robustness based on distance-constrained maximum likelihood , 2013, Comput. Stat. Data Anal..

[3]  Victor J. Yohai,et al.  Robust estimators for generalized linear models , 2014 .

[4]  Alfio Marazzi,et al.  Adaptively truncated maximum likelihood regression with asymmetric errors , 2004 .

[5]  A. Basu,et al.  Robust estimation in generalized linear models: the density power divergence approach , 2014, TEST.

[6]  Peter K. Dunn,et al.  Randomized Quantile Residuals , 1996 .

[7]  V. Yohai.  HIGH BREAKDOWN-POINT AND HIGH EFFICIENCY ROBUST ESTIMATES FOR REGRESSION , 1987 .

[8]  V. Yohai,et al.  High Breakdown-Point Estimates of Regression by Means of the Minimization of an Efficient Scale , 1988 .

[9]  V. Yohai,et al.  A class of robust and fully efficient regression estimators , 2002 .

[10]  P. J. Huber.  The 1972 Wald Lecture Robust Statistics: A Review , 1972 .

[11]  Aaron K. Han.  Non-parametric analysis of a generalized regression model: the maximum rank correlation estimator , 1987 .

[12]  Howard D Bondell,et al.  Efficient Robust Regression via Two-Stage Generalized Empirical Likelihood , 2013, Journal of the American Statistical Association.

[13]  Computational aspects of adaptive combination of least squares and least absolute deviations estimators , 1991 .

[14]  Pavel íek,et al.  Semiparametrically weighted robust estimation of regression models , 2011 .

[15]  P. Rousseeuw.  Multivariate estimation with high breakdown point , 1985 .

[16]  E. Ronchetti,et al.  Robust Inference for Generalized Linear Models , 2001 .

[17]  R. Carroll,et al.  Conditionally Unbiased Bounded-Influence Estimation in General Regression Models, with Applications to Generalized Linear Models , 1989 .

[18]  William H. Aeberhard,et al.  Robust inference in the negative binomial regression model with an application to falls data , 2014, Biometrics.

[19]  Peter Filzmoser,et al.  Robust joint modeling of mean and dispersion through trimming , 2012, Comput. Stat. Data Anal..