Box-constrained monotone approximations to Lipschitz regularizations, with applications to robust testing

DOI10.1007/S10957-020-01743-5zbMATH Open1466.62324arXiv1903.08573OpenAlexW2985190846MaRDI QIDQ831360FDOQ831360

Authors: Eustasio Del Barrio, Hristo Inouzhe, Carlos Matrán 

Publication date: 11 May 2021

Published in: Journal of Optimization Theory and Applications (Search for Journal in Brave)

Abstract: Tests of fit to exact models in statistical analysis often lead to rejections even when the model is a useful approximate description of the random generator of the data. Among possible relaxations of a fixed model, the one defined by contamination neighbourhoods, namely, $mathcal{V}_{a}lpha(P_0)=(1-alpha)P_0+alphaQ:QinmathcalP$, where $mathcalP$ is the set of all probabilities in the sample space, has received much attention, from its central role in Robust Statistics. For probabilities on the real line, consistent tests of fit to $mathcal{V}_{a}lpha(P_0)$ can be based on $d_K(P_0,R_{a}lpha(P))$, the minimal Kolmogorov distance between $P_0$ and the set of trimmings of $P$, $$. We show that this functional admits equivalent formulations in terms of, either best approximation in uniform norm by $L$-Lipschitz functions satisfying a box constraint, or as the best monotone approximation in uniform norm to the $L$-Lipschitz regularization, which is seen to be expressable in terms of the average of the Pasch-Hausdorff envelopes. This representation for the solution of the variational problem allows to obtain results showing stability of the functional $d_K(P_0,R_{a}lpha(P))$, as well as directional differentiability, providing the basis for a Central Limit Theorem for that functional.

Full work available at URL: https://arxiv.org/abs/1903.08573

Recommendations

• On approximate validation of models: a Kolmogorov-Smirnov-based approach
• scientific article; zbMATH DE number 19010
• scientific article; zbMATH DE number 5233411
• ROBUST SLIPPAGE TESTS II
• Robust test statsitics induced from the minimumL₂distance estimation

zbMATH Keywords

robustnessKolmogorov distancedistribution functiondirectional differentiabilityLipschitz regularizationcontamination neighbourhoodsLipschitz-continuous approximationsPasch-Hausdorff envelopestrimmed probabilities

Mathematics Subject Classification ID

Asymptotic properties of nonparametric inference (62G20) Nonparametric robustness (62G35) Lipschitz (Hölder) classes (26A16) Existence of optimal solutions to problems involving randomness (49J55) Approximation with constraints (41A29)

Cites Work

• Detecting relevant changes in the mean of nonstationary processes -- a mass excess approach
• Robust Estimation of a Location Parameter
• Estimating the proportion of false null hypotheses among a large number of independently tested hypotheses
• A stochastic process approach to false discovery control.
• Uniformly powerful goodness of fit tests
• Testing Statistical Hypotheses of Equivalence and Noninferiority
• Title not available (Why is that?)
• Best approximations to random variables based on trimming procedures
• Uniqueness and approximate computation of optimal incomplete transportation plans
• On concepts of directional differentiability
• Trimmed Comparison of Distributions
• Nonparametric Validation of Similar Distributions and Assessment of Goodness of Fit
• Similarity of samples and trimming
• Data features1
• Isotone optimization. II
• Isotone optimization. I
• Approximating data
• Title not available (Why is that?)
• Equivalence of regression curves
• Building and using semiparametric tolerance regions for parametric multinomial models
• Models for the assessment of treatment improvement: the ideal and the feasible
• Directional differentiability for supremum-type functionals: statistical applications
• A contamination model for the stochastic order
• On approximate validation of models: a Kolmogorov-Smirnov-based approach
• Detecting relevant changes in time series models
• Data analysis and approximate models. Model choice, location-scale, analysis of variance, nonparametic regression and image analysis
• Model assessment tools for a model false world

Cited In (1)

• On approximate validation of models: a Kolmogorov-Smirnov-based approach

This page was built for publication: Box-constrained monotone approximations to Lipschitz regularizations, with applications to robust testing

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q831360)