The null and non-null asymptotic distribution of the Cox test for multivariate nonlinear regression. Alternatives and a new distribution-free Cox test

Victor Aguirre-Torres, Andrew Ronald Gallant

Research output: Contribution to journalArticle

18 Citations (Scopus)

Abstract

The asymptotic distribution of the generalized Cox test for choosing between two multivariate, nonlinear regression models in implicit form is derived. The data is assumed to be generated by a model that need not be either the null or the non-null model. As the data-generating model is not subjected to a Pitman drift the analysis is global, not local, and provides a fairly complete qualitative description of the power characteristics of the generalized Cox test. Some investigations of these characteristics are included. A new test statistic is introduced that does not require an explicit specification of the error distribution of the null model. The idea is to replace an analytical computation of the expectation of the Cox difference with a bootstrap estimate. The null distribution of this new test is derived.

Original languageEnglish (US)
Pages (from-to)5-33
Number of pages29
JournalJournal of Econometrics
Volume21
Issue number1
DOIs
StatePublished - Jan 1 1983

Fingerprint

Multivariate Regression
Nonlinear Regression
Distribution-free
Asymptotic distribution
Null
Alternatives
Nonlinear Regression Model
Null Distribution
Model
Bootstrap
Test Statistic
Specification
Nonlinear regression
Estimate

All Science Journal Classification (ASJC) codes

  • Economics and Econometrics

Cite this

@article{03b67774af6f445a9cefee8778870495,
title = "The null and non-null asymptotic distribution of the Cox test for multivariate nonlinear regression. Alternatives and a new distribution-free Cox test",
abstract = "The asymptotic distribution of the generalized Cox test for choosing between two multivariate, nonlinear regression models in implicit form is derived. The data is assumed to be generated by a model that need not be either the null or the non-null model. As the data-generating model is not subjected to a Pitman drift the analysis is global, not local, and provides a fairly complete qualitative description of the power characteristics of the generalized Cox test. Some investigations of these characteristics are included. A new test statistic is introduced that does not require an explicit specification of the error distribution of the null model. The idea is to replace an analytical computation of the expectation of the Cox difference with a bootstrap estimate. The null distribution of this new test is derived.",
author = "Victor Aguirre-Torres and Gallant, {Andrew Ronald}",
year = "1983",
month = "1",
day = "1",
doi = "10.1016/0304-4076(83)90117-3",
language = "English (US)",
volume = "21",
pages = "5--33",
journal = "Journal of Econometrics",
issn = "0304-4076",
publisher = "Elsevier BV",
number = "1",

}

TY - JOUR

T1 - The null and non-null asymptotic distribution of the Cox test for multivariate nonlinear regression. Alternatives and a new distribution-free Cox test

AU - Aguirre-Torres, Victor

AU - Gallant, Andrew Ronald

PY - 1983/1/1

Y1 - 1983/1/1

N2 - The asymptotic distribution of the generalized Cox test for choosing between two multivariate, nonlinear regression models in implicit form is derived. The data is assumed to be generated by a model that need not be either the null or the non-null model. As the data-generating model is not subjected to a Pitman drift the analysis is global, not local, and provides a fairly complete qualitative description of the power characteristics of the generalized Cox test. Some investigations of these characteristics are included. A new test statistic is introduced that does not require an explicit specification of the error distribution of the null model. The idea is to replace an analytical computation of the expectation of the Cox difference with a bootstrap estimate. The null distribution of this new test is derived.

AB - The asymptotic distribution of the generalized Cox test for choosing between two multivariate, nonlinear regression models in implicit form is derived. The data is assumed to be generated by a model that need not be either the null or the non-null model. As the data-generating model is not subjected to a Pitman drift the analysis is global, not local, and provides a fairly complete qualitative description of the power characteristics of the generalized Cox test. Some investigations of these characteristics are included. A new test statistic is introduced that does not require an explicit specification of the error distribution of the null model. The idea is to replace an analytical computation of the expectation of the Cox difference with a bootstrap estimate. The null distribution of this new test is derived.

UR - http://www.scopus.com/inward/record.url?scp=0040313685&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0040313685&partnerID=8YFLogxK

U2 - 10.1016/0304-4076(83)90117-3

DO - 10.1016/0304-4076(83)90117-3

M3 - Article

AN - SCOPUS:0040313685

VL - 21

SP - 5

EP - 33

JO - Journal of Econometrics

JF - Journal of Econometrics

SN - 0304-4076

IS - 1

ER -