Probability inequalities in multivariate distributions

Research output: Contribution to journalArticle

5 Citations (Scopus)

Abstract

For a bivariate binary response model y, = 1 (xj βj+j > 0), j=1,2, we propose to estimate nonpararnetrically the quadrant correlation E{sgn(u1) *sgn(u2)} between the two error terms ul and u2 without specifjing the error term distribution. The quadrant correlation accounts for the relationship between yl and y2 that is not explained by xl and x2, and can be used in testing for the specification of endogenous dummy variable models. The quadrant correlation is further generalized into orthant dependence allowing unknown regression functions, unknown error term distribution and arbitrary forms of heteroskedasticity. A simulation study is provided, followed by a brief application to a real data set.

Original languageEnglish
Pages (from-to)387-415
Number of pages29
JournalEconometric Reviews
Volume18
Issue number4
DOIs
Publication statusPublished - 1999 Jan 1

Fingerprint

Multivariate distribution
Binary response model
Heteroskedasticity
Dummy variables
Testing
Simulation study

Keywords

  • Binary response
  • Endogenous dummy varible
  • Orthant dependence
  • Quadrant correlation

ASJC Scopus subject areas

  • Economics and Econometrics

Cite this

Probability inequalities in multivariate distributions. / Lee, Myoung-jae.

In: Econometric Reviews, Vol. 18, No. 4, 01.01.1999, p. 387-415.

Research output: Contribution to journalArticle

@article{d4acd50db1134653aed8db8c89617133,
title = "Probability inequalities in multivariate distributions",
abstract = "For a bivariate binary response model y, = 1 (xj βj+j > 0), j=1,2, we propose to estimate nonpararnetrically the quadrant correlation E{sgn(u1) *sgn(u2)} between the two error terms ul and u2 without specifjing the error term distribution. The quadrant correlation accounts for the relationship between yl and y2 that is not explained by xl and x2, and can be used in testing for the specification of endogenous dummy variable models. The quadrant correlation is further generalized into orthant dependence allowing unknown regression functions, unknown error term distribution and arbitrary forms of heteroskedasticity. A simulation study is provided, followed by a brief application to a real data set.",
keywords = "Binary response, Endogenous dummy varible, Orthant dependence, Quadrant correlation",
author = "Myoung-jae Lee",
year = "1999",
month = "1",
day = "1",
doi = "10.1080/07474939908800352",
language = "English",
volume = "18",
pages = "387--415",
journal = "Econometric Reviews",
issn = "0747-4938",
publisher = "Taylor and Francis Ltd.",
number = "4",

}

TY - JOUR

T1 - Probability inequalities in multivariate distributions

AU - Lee, Myoung-jae

PY - 1999/1/1

Y1 - 1999/1/1

N2 - For a bivariate binary response model y, = 1 (xj βj+j > 0), j=1,2, we propose to estimate nonpararnetrically the quadrant correlation E{sgn(u1) *sgn(u2)} between the two error terms ul and u2 without specifjing the error term distribution. The quadrant correlation accounts for the relationship between yl and y2 that is not explained by xl and x2, and can be used in testing for the specification of endogenous dummy variable models. The quadrant correlation is further generalized into orthant dependence allowing unknown regression functions, unknown error term distribution and arbitrary forms of heteroskedasticity. A simulation study is provided, followed by a brief application to a real data set.

AB - For a bivariate binary response model y, = 1 (xj βj+j > 0), j=1,2, we propose to estimate nonpararnetrically the quadrant correlation E{sgn(u1) *sgn(u2)} between the two error terms ul and u2 without specifjing the error term distribution. The quadrant correlation accounts for the relationship between yl and y2 that is not explained by xl and x2, and can be used in testing for the specification of endogenous dummy variable models. The quadrant correlation is further generalized into orthant dependence allowing unknown regression functions, unknown error term distribution and arbitrary forms of heteroskedasticity. A simulation study is provided, followed by a brief application to a real data set.

KW - Binary response

KW - Endogenous dummy varible

KW - Orthant dependence

KW - Quadrant correlation

UR - http://www.scopus.com/inward/record.url?scp=22944475344&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=22944475344&partnerID=8YFLogxK

U2 - 10.1080/07474939908800352

DO - 10.1080/07474939908800352

M3 - Article

VL - 18

SP - 387

EP - 415

JO - Econometric Reviews

JF - Econometric Reviews

SN - 0747-4938

IS - 4

ER -