A loss function approach to model specification testing and its relative efficiency
From MaRDI portal
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.
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
Cites work
- scientific article; zbMATH DE number 3143211 (Why is no real title available?)
- scientific article; zbMATH DE number 3690440 (Why is no real title available?)
- scientific article; zbMATH DE number 48302 (Why is no real title available?)
- scientific article; zbMATH DE number 472958 (Why is no real title available?)
- scientific article; zbMATH DE number 1044059 (Why is no real title available?)
- scientific article; zbMATH DE number 1168350 (Why is no real title available?)
- A CONSISTENT MODEL SPECIFICATION TEST BASED ON THE KERNEL SUM OF SQUARES OF RESIDUALS
- A Look at Some Data on the Old Faithful Geyser
- A Reexamination of Diffusion Estimators With Applications to Financial Model Validation
- A loss function approach to model specification testing and its relative efficiency
- A power comparison between nonparametric regression tests.
- An Adaptive, Rate-Optimal Test of a Parametric Mean-Regression Model Against a Nonparametric Alternative
- Approximation Theorems of Mathematical Statistics
- Asymptotically minimax hypothesis testing for nonparametric alternatives. I
- Asymptotically minimax hypothesis testing for nonparametric alternatives. II
- Asymptotically minimax hypothesis testing for nonparametric alternatives. III
- Autoregressive Conditional Heteroscedasticity with Estimates of the Variance of United Kingdom Inflation
- Bandwidth selection in nonparametric kernel testing
- Bayesian Estimation and Prediction Using Asymmetric Loss Functions
- Bootstrap of kernel smoothing in nonlinear time series
- Comparing nonparametric versus parametric regression fits
- Consistent Nonparametric Entropy-Based Testing
- Consistent Specification Testing Via Nonparametric Series Regression
- Decision-based methods for forecast evaluation
- Econometric Model Determination
- Functional-Coefficient Regression Models for Nonlinear Time Series
- Generalised likelihood ratio tests for spectral density
- Generalized Spectral Tests for Conditional Mean Models in Time Series with Conditional Heteroscedasticity of Unknown Form
- Generalized likelihood ratio statistics and Wilks phenomenon
- Likelihood Ratio Tests for Model Selection and Non-Nested Hypotheses
- Maximum Likelihood Estimation of Misspecified Models
- Maximum Likelihood: An Introduction
- Minimax nonparametric hypothesis testing: The case of an inhomogeneous alternative
- Mixing: Properties and examples
- Nonlinear time series. Nonparametric and parametric methods
- Nonparametric Inferences for Additive Models
- Nonparametric econometrics. Theory and practice.
- Nonparametric inference with generalized likelihood ratio tests (With comments and rejoinder)
- Nonparametric statistics for testing of linearity and serial independence
- On the use of nonparametric regression for model checking
- Optimal Bandwidth Selection in Heteroskedasticity–Autocorrelation Robust Testing
- Optimal monetary policy with a nonlinear Phillips curve
- Profile likelihood inferences on semiparametric varying-coefficient partially linear models
- Testing for neglected nonlinearity in time series models. A comparison of neural network methods and alternative tests
- Tests of Conditional Predictive Ability
- Weighted least absolute deviations estimation for ARMA models with infinite variance
Cited in
(9)- A loss function related to the FDR for random effects multiple comparisons
- False discovery rates for large-scale model checking under certain dependence
- Goodness-of-fit testing of error distribution in linear measurement error models
- Multidimensional specification test based on non-stationary time series
- Nonparametric inference for additive models estimated via simplified smooth backfitting
- A loss function approach to model specification testing and its relative efficiency
- Multiple-criteria fuzzy group decision-making with multi-choice goal programming for supplier selection: a case study
- Tests and estimation strategies associated to some loss functions
- A nonparametric test of significant variables in gradients
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)
