Empirical Likelihood Ratio Test With Distribution Function Constraints

DOI10.1109/TSP.2013.2271484zbMATH Open1393.94590arXiv1605.00157OpenAlexW2088475901MaRDI QIDQ4578722FDOQ4578722

Authors: Yingxi Liu, Ahmed H. Tewfik 

Publication date: 22 August 2018

Published in: IEEE Transactions on Signal Processing (Search for Journal in Brave)

Abstract: In this work, we study non-parametric hypothesis testing problem with distribution function constraints. The empirical likelihood ratio test has been widely used in testing problems with moment (in)equality constraints. However, some detection problems cannot be described using moment (in)equalities. We propose a distribution function constraint along with an empirical likelihood ratio test. This detector is applicable to a wide variety of robust parametric/non-parametric detection problems. Since the distribution function constraints provide a more exact description of the null hypothesis, the test outperforms the empirical likelihood ratio test with moment constraints as well as many popular goodness-of-fit tests, such as the robust Kolmogorov-Smirnov test and the Cram'er-von Mises test. Examples from communication systems with real-world noise samples are provided to show their performance. Specifically, the proposed test significantly outperforms the robust Kolmogorov-Smirnov test and the Cram'er-von Mises test when the null hypothesis is nested in the alternative hypothesis. The same example is repeated when we assume no noise uncertainty. By doing so, we are able to claim that in our case, it is necessary to include uncertainty in noise distribution. Additionally, the asymptotic optimality of the proposed test is provided.

Full work available at URL: https://arxiv.org/abs/1605.00157

Recommendations

• Empirical likelihood estimation and consistent tests with conditional moment restrictions
• Empirical likelihood hypothesis test on mean with inequality constraints
• Empirical likelihood ratio test on quantiles under a density ratio model
• Empirical likelihood test via estimating equations
• Likelihood ratio test against a set of inequality constraints
• Tests de la razón de verosimilitud para medias de poblaciones normales, sujetas a restricciones
• Likelihood ratio tests with boundary constraints using data-dependent degrees of freedom
• A consistent jackknife empirical likelihood test for distribution functions

Mathematics Subject Classification ID

Nonparametric hypothesis testing (62G10) Nonparametric robustness (62G35) Detection theory in information and communication theory (94A13)

Cited In (4)

• Dominance of likelihood ratio tests under cone constraints
• A Multi-level Refinement Approach for Structural Synthesis of Optimal Probabilistic Models
• Tests de la razón de verosimilitud para medias de poblaciones normales, sujetas a restricciones
• Title not available (Why is that?)

This page was built for publication: Empirical Likelihood Ratio Test With Distribution Function Constraints

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4578722)