Orthogonality of the mean and error distribution in generalized linear models

DOI10.1080/03610926.2013.851241MaRDI QIDQ4976210zbMATH OpenOpenAlexWikidataFDO

Authors Alan Huang, Paul J. Rathouz

Publication date 27 July 2017

Published in Communications in Statistics: Theory and Methods (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1606.01604, http://europepmc.org/articles/pmc5396964

zbMATH Keywords

nuisance tangent spaceregression modelsemiparametric model

Mathematics Subject Classification ID

Generalized linear models (logistic models) (62J12)

Abstract: We show that the mean-model parameter is always orthogonal to the error distribution in generalized linear models. Thus, the maximum likelihood estimator of the mean-model parameter will be asymptotically efficient regardless of whether the error distribution is known completely, known up to a finite vector of parameters, or left completely unspecified, in which case the likelihood is taken to be an appropriate semiparametric likelihood. Moreover, the maximum likelihood estimator of the mean-model parameter will be asymptotically independent of the maximum likelihood estimator of the error distribution. This generalizes some well-known results for the special cases of normal, gamma and multinomial regression models, and, perhaps more interestingly, suggests that asymptotically efficient estimation and inferences can always be obtained if the error distribution is nonparametrically estimated along with the mean. In contrast, estimation and inferences using misspecified error distributions or variance functions are generally not efficient.

Recommendations

• Maximum likelihood estimation in misspecified generalized linear models
• A mote on mle's in linear models with non-normal or dependent errors
• On selecting parametric link transformation families in generalized linear models
• The maximum likelihood estimator for joint mean and dispersion in generalized linear models of the lognormal distribution
• Measurement error in the generalised linear model

Cites work

• scientific article; zbMATH DE number 3945130 (Why is no real title available?)
• INTERPRETATION OF THE QUASI-LIKELIHOOD VIA THE TILTED EXPONENTIAL FAMILY
• Parameter Orthogonality and Bias Adjustment for Estimating Functions
• Semiparametric theory and missing data.

Cited in

(8)

• Inference in Error Orthogonal Models
• Mean-parametrized Conway–Maxwell–Poisson regression models for dispersed counts
• Smooth tests of goodness of fit for the distributional assumption of regression models
• Generalized case‐control sampling under generalized linear models
• A fast look-up method for Bayesian mean-parameterised Conway-Maxwell-Poisson regression models
• Finite mixtures of mean-parameterized Conway-Maxwell-Poisson regressions
• Finite mixtures of mean-parameterized Conway-Maxwell-Poisson models
• On exponential-family INGARCH models

This page was built for publication: Orthogonality of the mean and error distribution in generalized linear models

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4976210)