Optimal comparison of misspecified moment restriction models under a chosen measure of fit

Vadim Marmer, Taisuke Otsu

Research output: Contribution to journalArticle

3 Citations (Scopus)

Abstract

Suppose that the econometrician is interested in comparing two misspecified moment restriction models, where the comparison is performed in terms of some chosen measure of fit. This paper is concerned with describing an optimal test of the Vuong (1989) and Rivers and Vuong (2002) type null hypothesis that the two models are equivalent under the given measure of fit (the ranking may vary for different measures). We adopt the generalized NeymanPearson optimality criterion, which focuses on the decay rates of the type I and II error probabilities under fixed non-local alternatives, and derive an optimal but practically infeasible test. Then, as an illustration, by considering the model comparison hypothesis defined by the weighted Euclidean norm of moment restrictions, we propose a feasible approximate test statistic to the optimal one and study its asymptotic properties. Local power properties, one-sided test, and comparison under the generalized empirical likelihood-based measure of fit are also investigated. A simulation study illustrates that our approximate test is more powerful than the RiversVuong test.

Original languageEnglish
Pages (from-to)538-550
Number of pages13
JournalJournal of Econometrics
Volume170
Issue number2
DOIs
Publication statusPublished - 2012 Oct 1
Externally publishedYes

Fingerprint

Restriction
Moment
One-sided test
Local Power
Type II error
Optimal Test
Weighted Norm
Euclidean norm
Model Comparison
Empirical Likelihood
Type I error
Optimality Criteria
Error Probability
Rivers
Decay Rate
Null hypothesis
Statistics
Model
Asymptotic Properties
Test Statistic

Keywords

  • Generalized method of moments
  • Generalized NeymanPearson optimality
  • Misspecification
  • Model comparison
  • Moment restriction

ASJC Scopus subject areas

  • Economics and Econometrics
  • Applied Mathematics
  • History and Philosophy of Science

Cite this

Optimal comparison of misspecified moment restriction models under a chosen measure of fit. / Marmer, Vadim; Otsu, Taisuke.

In: Journal of Econometrics, Vol. 170, No. 2, 01.10.2012, p. 538-550.

Research output: Contribution to journalArticle

@article{f81c58300bf24cbc8c22c02003b6b61a,
title = "Optimal comparison of misspecified moment restriction models under a chosen measure of fit",
abstract = "Suppose that the econometrician is interested in comparing two misspecified moment restriction models, where the comparison is performed in terms of some chosen measure of fit. This paper is concerned with describing an optimal test of the Vuong (1989) and Rivers and Vuong (2002) type null hypothesis that the two models are equivalent under the given measure of fit (the ranking may vary for different measures). We adopt the generalized NeymanPearson optimality criterion, which focuses on the decay rates of the type I and II error probabilities under fixed non-local alternatives, and derive an optimal but practically infeasible test. Then, as an illustration, by considering the model comparison hypothesis defined by the weighted Euclidean norm of moment restrictions, we propose a feasible approximate test statistic to the optimal one and study its asymptotic properties. Local power properties, one-sided test, and comparison under the generalized empirical likelihood-based measure of fit are also investigated. A simulation study illustrates that our approximate test is more powerful than the RiversVuong test.",
keywords = "Generalized method of moments, Generalized NeymanPearson optimality, Misspecification, Model comparison, Moment restriction",
author = "Vadim Marmer and Taisuke Otsu",
year = "2012",
month = "10",
day = "1",
doi = "10.1016/j.jeconom.2012.05.021",
language = "English",
volume = "170",
pages = "538--550",
journal = "Journal of Econometrics",
issn = "0304-4076",
publisher = "Elsevier BV",
number = "2",

}

TY - JOUR

T1 - Optimal comparison of misspecified moment restriction models under a chosen measure of fit

AU - Marmer, Vadim

AU - Otsu, Taisuke

PY - 2012/10/1

Y1 - 2012/10/1

N2 - Suppose that the econometrician is interested in comparing two misspecified moment restriction models, where the comparison is performed in terms of some chosen measure of fit. This paper is concerned with describing an optimal test of the Vuong (1989) and Rivers and Vuong (2002) type null hypothesis that the two models are equivalent under the given measure of fit (the ranking may vary for different measures). We adopt the generalized NeymanPearson optimality criterion, which focuses on the decay rates of the type I and II error probabilities under fixed non-local alternatives, and derive an optimal but practically infeasible test. Then, as an illustration, by considering the model comparison hypothesis defined by the weighted Euclidean norm of moment restrictions, we propose a feasible approximate test statistic to the optimal one and study its asymptotic properties. Local power properties, one-sided test, and comparison under the generalized empirical likelihood-based measure of fit are also investigated. A simulation study illustrates that our approximate test is more powerful than the RiversVuong test.

AB - Suppose that the econometrician is interested in comparing two misspecified moment restriction models, where the comparison is performed in terms of some chosen measure of fit. This paper is concerned with describing an optimal test of the Vuong (1989) and Rivers and Vuong (2002) type null hypothesis that the two models are equivalent under the given measure of fit (the ranking may vary for different measures). We adopt the generalized NeymanPearson optimality criterion, which focuses on the decay rates of the type I and II error probabilities under fixed non-local alternatives, and derive an optimal but practically infeasible test. Then, as an illustration, by considering the model comparison hypothesis defined by the weighted Euclidean norm of moment restrictions, we propose a feasible approximate test statistic to the optimal one and study its asymptotic properties. Local power properties, one-sided test, and comparison under the generalized empirical likelihood-based measure of fit are also investigated. A simulation study illustrates that our approximate test is more powerful than the RiversVuong test.

KW - Generalized method of moments

KW - Generalized NeymanPearson optimality

KW - Misspecification

KW - Model comparison

KW - Moment restriction

UR - http://www.scopus.com/inward/record.url?scp=84865330387&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84865330387&partnerID=8YFLogxK

U2 - 10.1016/j.jeconom.2012.05.021

DO - 10.1016/j.jeconom.2012.05.021

M3 - Article

VL - 170

SP - 538

EP - 550

JO - Journal of Econometrics

JF - Journal of Econometrics

SN - 0304-4076

IS - 2

ER -