Mean and median bias reduction in generalized linear models

DOI10.1007/S11222-019-09860-6zbMATH Open1436.62351arXiv1804.04085OpenAlexW2963667546WikidataQ128289889 ScholiaQ128289889MaRDI QIDQ77213FDOQ77213

Authors: Euloge Clovis Kenne Pagui, Nicola Sartori, Ioannis Kosmidis, Euloge Clovis Kenne Pagui, N. Sartori 

Publication date: 4 March 2019

Published in: Statistics and Computing (Search for Journal in Brave)

Abstract: This paper presents an integrated framework for estimation and inference from generalized linear models using adjusted score equations that result in mean and median bias reduction. The framework unifies theoretical and methodological aspects of past research on mean bias reduction and accommodates, in a natural way, new advances on median bias reduction. General expressions for the adjusted score functions are derived in terms of quantities that are readily available in standard software for fitting generalized linear models. The resulting estimating equations are solved using a unifying quasi-Fisher scoring algorithm that is shown to be equivalent to iteratively re-weighted least squares with appropriately adjusted working variates. Formal links between the iterations for mean and median bias reduction are established. Core model invariance properties are used to develop a novel mixed adjustment strategy when the estimation of a dispersion parameter is necessary. It is also shown how median bias reduction in multinomial logistic regression can be done using the equivalent Poisson log-linear model. The estimates coming out from mean and median bias reduction are found to overcome practical issues related to infinite estimates that can occur with positive probability in generalized linear models with multinomial or discrete responses, and can result in valid inferences even in the presence of a high-dimensional nuisance parameter

Full work available at URL: https://arxiv.org/abs/1804.04085

Recommendations

• Median bias reduction of maximum likelihood estimates
• Median bias reduction in cumulative link models
• On bias reduction in robust inference for generalized linear models
• scientific article; zbMATH DE number 469130
• Bias in Multivariate Generalized Linear Models
• Median based covariogram estimators reduce bias
• Simulation methods for mean and median bias reduction in parametric estimation
• Bias reduction for nonparametric and semiparametric regression models
• Bias reduction in exponential family nonlinear models

zbMATH Keywords

adjusted score equationsdata separationdispersioniterative reweighted least squaresmultinomial regressionparameterization invariance

Mathematics Subject Classification ID

Point estimation (62F10) General nonlinear regression (62J02) Generalized linear models (logistic models) (62J12)

Cites Work

• Bias reduction in exponential family nonlinear models
• Inference functions and quadratic score tests
• Title not available (Why is that?)
• Defining the curvature of a statistical problem (with applications to second order efficiency)
• A modified score function estimator for multinomial logistic regression in small samples.
• A generic algorithm for reducing bias in parametric estimation
• Modified profile likelihoods in models with stratum nuisance parameters
• Title not available (Why is that?)
• Multinomial logit bias reduction via the Poisson log-linear model
• Title not available (Why is that?)
• Title not available (Why is that?)
• On a formula for the distribution of the maximum likelihood estimator
• On the existence of maximum likelihood estimates in logistic regression models
• On Bayesian Analysis of Generalized Linear Models Using Jeffreys's Prior
• On the existence and uniqueness of the maximum likelihood estimates for certain generalized linear models
• An approximation to the modified profile likelihood function
• Title not available (Why is that?)
• Reducing bias and mean squared error associated with regression-based odds ratio estimators
• Title not available (Why is that?)
• Bias reduction of maximum likelihood estimates
• Title not available (Why is that?)
• Median bias reduction of maximum likelihood estimates
• On bias reduction and incidental parameters
• Applied Asymptotics
• Title not available (Why is that?)
• Improved Estimation in Cumulative Link Models
• Jeffreys-prior penalty, finiteness and shrinkage in binomial-response generalized linear models

Cited In (14)

• Adjusted score functions for monotone likelihood in the Cox regression model
• Bias reduction of maximum likelihood estimates for an asymmetric class of power models with applications
• Bayesian Lesion Estimation with a Structured Spike-and-Slab Prior
• Bounded-memory adjusted scores estimation in generalized linear models with large data sets
• A general expression for second-order covariance matrices -- an application to dispersion models
• A comprehensive review of bias reduction methods for logistic regression
• Conjugate priors and bias reduction for logistic regression models
• Bias‐reduced estimators of conditional odds ratios in matched case‐control studies with unmatched confounding
• Properties of Fisher information gain for Bayesian design of experiments
• Improved estimation in negative binomial regression
• brglm2
• Bias reduction in exponential family nonlinear models
• Integration of model-based recursive partitioning with bias reduction estimation: a case study assessing the impact of Oliver's four factors on the probability of winning a basketball game
• Median bias reduction in cumulative link models

Uses Software

• R
• brglm2

This page was built for publication: Mean and median bias reduction in generalized linear models

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q77213)