M-estimation for discrete data: Asymptotic distribution theory and implications

DOI10.1214/AOS/1176350367MaRDI QIDQ1107231zbMATH OpenOpenAlexFDO

Authors Douglas G. Simpson, David Ruppert, Raymond J. Carroll

Publication date 1987

Published in The Annals of Statistics (Search for Journal in Brave)

Full work available at URL https://doi.org/10.1214/aos/1176350367

zbMATH Keywords

asymptotic normalitydiscrete modelM-estimatornon-normal limiting distributionsrobust estimationsmooth score functions

Mathematics Subject Classification ID

Asymptotic distribution theory in statistics (62E20) Point estimation (62F10) Robustness and adaptive procedures (parametric inference) (62F35)

Review text

The authors consider the asymptotic distribution of an M-estimator when the underlying distribution is discrete. They show its asymptotic normality quite generally. When the specification of the model is inexact, however, non-normal limiting distributions may result, and lead the authors to propose smooth score functions for discrete data.

Recommendations

• A second-order asymptotic distributional representation of M-estimators with discontinuous score functions
• Asymptotic Theory for M-Estimators of Boundaries
• Estimation in discrete parameter models
• A general asymptotic theory of \(M\)-estimators. II
• M-estimation in linear models under nonstandard conditions.

Cited in

(11)

• Weighted likelihood estimating equations: The discrete case with applications to logistic regression
• Local M-estimation with discontinuous criterion for dependent and limited observations
• Robust fitting of the binomial model.
• M estimators based on the probability integral transformation with applications to count data
• Robust statistical inference based on the \(C\)-divergence family
• Properties of robust m-estimators for poisson and negative binomial data^∗
• scientific article; zbMATH DE number 4022397 (Why is no real title available?)
• Robust fitting of INARCH models
• Estimation in discrete parameter models
• Recursive estimators for stationary, strong mixing processes - a representation theorem and asymptotic distributions
• Trimmed and Winsorized means based on a scaled deviation

This page was built for publication: M-estimation for discrete data: Asymptotic distribution theory and implications

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1107231)