A loss function approach to model specification testing and its relative efficiency

DOI10.1214/13-AOS1099zbMATH Open1293.62100arXiv1306.4864MaRDI QIDQ366964FDOQ366964

Authors: Yongmiao Hong, Yoon-Jin Lee 

Publication date: 25 September 2013

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

Abstract: The generalized likelihood ratio (GLR) test proposed by Fan, Zhang and Zhang [Ann. Statist. 29 (2001) 153-193] and Fan and Yao [Nonlinear Time Series: Nonparametric and Parametric Methods (2003) Springer] is a generally applicable nonparametric inference procedure. In this paper, we show that although it inherits many advantages of the parametric maximum likelihood ratio (LR) test, the GLR test does not have the optimal power property. We propose a generally applicable test based on loss functions, which measure discrepancies between the null and nonparametric alternative models and are more relevant to decision-making under uncertainty. The new test is asymptotically more powerful than the GLR test in terms of Pitman's efficiency criterion. This efficiency gain holds no matter what smoothing parameter and kernel function are used and even when the true likelihood function is available for the GLR test.

Full work available at URL: https://arxiv.org/abs/1306.4864

Recommendations

• A predictive motivation for loss function specification in parametric hypothesis testing
• Tests and estimation strategies associated to some loss functions
• Robust and efficient fitting of loss models
• Weighted Loss Functions for Set Estimation and Testing Hypotheses
• Exact likelihood ratio scale and homogeneity testing of some loss processes
• The Loss Rank Principle for Model Selection
• Estimation based on an appropriate choice of loss function
• Model selection with the loss rank principle
• TESTING MODEL SPECIFICATION IN SEEMINGLY UNRELATED REGRESSION MODELS

zbMATH Keywords

efficiencykernelsmoothing parametergeneralized likelihood ratio testloss functionlocal alternativePitman efficiency

Mathematics Subject Classification ID

Nonparametric hypothesis testing (62G10)

Cites Work

• Approximation Theorems of Mathematical Statistics
• Maximum Likelihood Estimation of Misspecified Models
• Likelihood Ratio Tests for Model Selection and Non-Nested Hypotheses
• Mixing: Properties and examples
• Nonparametric econometrics. Theory and practice.
• Autoregressive Conditional Heteroscedasticity with Estimates of the Variance of United Kingdom Inflation
• Comparing nonparametric versus parametric regression fits
• Bootstrap of kernel smoothing in nonlinear time series
• Generalized likelihood ratio statistics and Wilks phenomenon
• Nonlinear time series. Nonparametric and parametric methods
• Bayesian Estimation and Prediction Using Asymmetric Loss Functions
• Title not available (Why is that?)
• Title not available (Why is that?)
• An Adaptive, Rate-Optimal Test of a Parametric Mean-Regression Model Against a Nonparametric Alternative
• Consistent Specification Testing Via Nonparametric Series Regression
• Profile likelihood inferences on semiparametric varying-coefficient partially linear models
• Tests of Conditional Predictive Ability
• A Look at Some Data on the Old Faithful Geyser
• Functional-Coefficient Regression Models for Nonlinear Time Series
• Optimal Bandwidth Selection in Heteroskedasticity–Autocorrelation Robust Testing
• Minimax nonparametric hypothesis testing: The case of an inhomogeneous alternative
• Asymptotically minimax hypothesis testing for nonparametric alternatives. I
• Asymptotically minimax hypothesis testing for nonparametric alternatives. II
• Title not available (Why is that?)
• Bandwidth selection in nonparametric kernel testing
• Testing for neglected nonlinearity in time series models. A comparison of neural network methods and alternative tests
• Decision-based methods for forecast evaluation
• Maximum Likelihood: An Introduction
• Title not available (Why is that?)
• Generalized Spectral Tests for Conditional Mean Models in Time Series with Conditional Heteroscedasticity of Unknown Form
• Consistent Nonparametric Entropy-Based Testing
• A power comparison between nonparametric regression tests.
• Title not available (Why is that?)
• A Reexamination of Diffusion Estimators With Applications to Financial Model Validation
• Generalised likelihood ratio tests for spectral density
• Weighted least absolute deviations estimation for ARMA models with infinite variance
• Nonparametric Inferences for Additive Models
• Econometric Model Determination
• Asymptotically minimax hypothesis testing for nonparametric alternatives. III
• Nonparametric inference with generalized likelihood ratio tests (With comments and rejoinder)
• On the use of nonparametric regression for model checking
• Nonparametric statistics for testing of linearity and serial independence
• Optimal monetary policy with a nonlinear Phillips curve
• Title not available (Why is that?)
• A loss function approach to model specification testing and its relative efficiency
• A CONSISTENT MODEL SPECIFICATION TEST BASED ON THE KERNEL SUM OF SQUARES OF RESIDUALS

Cited In (9)

• Tests and estimation strategies associated to some loss functions
• False discovery rates for large-scale model checking under certain dependence
• A loss function related to the FDR for random effects multiple comparisons
• Goodness-of-fit testing of error distribution in linear measurement error models
• A nonparametric test of significant variables in gradients
• A loss function approach to model specification testing and its relative efficiency
• Multidimensional specification test based on non-stationary time series
• Multiple-criteria fuzzy group decision-making with multi-choice goal programming for supplier selection: a case study
• Nonparametric inference for additive models estimated via simplified smooth backfitting

This page was built for publication: A loss function approach to model specification testing and its relative efficiency

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q366964)