Robust estimation on a parametric model via testing

DOI10.3150/15-BEJ706MaRDI QIDQ282553zbMATH OpenOpenAlexFDO

Authors Mathieu Sart

Publication date 12 May 2016

Published in Bernoulli (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1308.2927, https://projecteuclid.org/euclid.bj/1458132994

zbMATH Keywords

parametric estimationrobust estimationrobust testsT-estimator

Mathematics Subject Classification ID

Parametric hypothesis testing (62F03) Point estimation (62F10) Asymptotic properties of parametric estimators (62F12) Robustness and adaptive procedures (parametric inference) (62F35)

Abstract: We are interested in the problem of robust parametric estimation of a density from $n$ i.i.d. observations. By using a practice-oriented procedure based on robust tests, we build an estimator for which we establish non-asymptotic risk bounds with respect to the Hellinger distance under mild assumptions on the parametric model. We show that the estimator is robust even for models for which the maximum likelihood method is bound to fail. A numerical simulation illustrates its robustness properties. When the model is true and regular enough, we prove that the estimator is very close to the maximum likelihood one, at least when the number of observations $n$ is large. In particular, it inherits its efficiency. Simulations show that these two estimators are almost equal with large probability, even for small values of $n$ when the model is regular enough and contains the true density.

Recommendations

• A new method for estimation and model selection: \(\rho\)-estimation
• Robust and efficient estimation by minimising a density power divergence
• On minimum Hellinger distance estimation
• On robust estimation via pseudo-additive information
• Robust Bounded-Influence Tests in General Parametric Models

Cites work

• scientific article; zbMATH DE number 3839126 (Why is no real title available?)
• scientific article; zbMATH DE number 3868406 (Why is no real title available?)
• scientific article; zbMATH DE number 1191514 (Why is no real title available?)
• scientific article; zbMATH DE number 3690440 (Why is no real title available?)
• scientific article; zbMATH DE number 3602488 (Why is no real title available?)
• A general estimation method using spacings
• A new method for estimation and model selection: \(\rho\)-estimation
• AN ALTERNATIVE TO MAXIMUM LIKELIHOOD BASED ON SPACINGS
• An Inconsistent Maximum Likelihood Estimate
• Efficiency versus robustness: The case for minimum Hellinger distance and related methods
• Estimating the intensity of a random measure by histogram type estimators
• Estimation of the density of a determinantal process
• Estimation of the transition density of a Markov chain
• Estimator selection with respect to Hellinger-type risks
• Evaluation of the maximum-likelihood estimator where the likelihood equation has multiple roots
• Maximum Likelihood: An Introduction
• Minimum Hellinger distance estimates for parametric models
• Model selection for Poisson processes
• Model selection for Poisson processes with covariates
• Model selection for density estimation with \(\mathbb L_2\)-loss
• Model selection via testing: an alternative to (penalized) maximum likelihood estimators.
• On the consistency of the maximum spacing method
• Robust tests for model selection
• Statistical Inference
• Strong consistency of the maximum product of spacings estimates with applications in nonparametrics and in estimation of unimodal densities
• The ``automatic robustness of minimum distance functionals

Cited in

(7)

• A new method for estimation and model selection: \(\rho\)-estimation
• scientific article; zbMATH DE number 4143273 (Why is no real title available?)
• Parametric robust test for several variances with unknown underlying distributions
• Rendering parametric procedures more robust by empirically tilting the model
• Robust tests for model selection
• On robust sequential parameters estimating
• scientific article; zbMATH DE number 7383840 (Why is no real title available?)

This page was built for publication: Robust estimation on a parametric model via testing

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q282553)