Approximating the distribution of the Durbin-Watson statistic

Dennis P. Sheehan; Dennis P. Sheehan

doi:10.1080/03610928608829107

Approximating the distribution of the Durbin-Watson statistic

Dennis P. Sheehan, Dennis P. Sheehan

Finance

Research output: Contribution to journal › Article › peer-review

2 Scopus citations

Abstract

This paper derives approximations to the distribution of the Durbin-Watson statistic. When the coefficient of serial correlation is zero, the approximations are as good as other approximations proposed in the literature and probably somewhat easier to use. When serial correlation is present, the approximations are only moderately accurate. The problem is that the Edgeworth expansions derived here have difficulty replicating the shape of the true distribution function.

Original language	English (US)
Pages (from-to)	73-88
Number of pages	16
Journal	Communications in Statistics - Theory and Methods
Volume	15
Issue number	1
DOIs	https://doi.org/10.1080/03610928608829107
State	Published - Jan 1 1986

All Science Journal Classification (ASJC) codes

Statistics and Probability

Access to Document

10.1080/03610928608829107

Cite this

@article{b9781ab77b6f4ebf97f2ee5af5093345,

title = "Approximating the distribution of the Durbin-Watson statistic",

abstract = "This paper derives approximations to the distribution of the Durbin-Watson statistic. When the coefficient of serial correlation is zero, the approximations are as good as other approximations proposed in the literature and probably somewhat easier to use. When serial correlation is present, the approximations are only moderately accurate. The problem is that the Edgeworth expansions derived here have difficulty replicating the shape of the true distribution function.",

author = "Sheehan, {Dennis P.} and Sheehan, {Dennis P.}",

year = "1986",

month = jan,

day = "1",

doi = "10.1080/03610928608829107",

language = "English (US)",

volume = "15",

pages = "73--88",

journal = "Communications in Statistics - Theory and Methods",

issn = "0361-0926",

publisher = "Taylor and Francis Ltd.",

number = "1",

}

TY - JOUR

T1 - Approximating the distribution of the Durbin-Watson statistic

AU - Sheehan, Dennis P.

PY - 1986/1/1

Y1 - 1986/1/1

N2 - This paper derives approximations to the distribution of the Durbin-Watson statistic. When the coefficient of serial correlation is zero, the approximations are as good as other approximations proposed in the literature and probably somewhat easier to use. When serial correlation is present, the approximations are only moderately accurate. The problem is that the Edgeworth expansions derived here have difficulty replicating the shape of the true distribution function.

AB - This paper derives approximations to the distribution of the Durbin-Watson statistic. When the coefficient of serial correlation is zero, the approximations are as good as other approximations proposed in the literature and probably somewhat easier to use. When serial correlation is present, the approximations are only moderately accurate. The problem is that the Edgeworth expansions derived here have difficulty replicating the shape of the true distribution function.

UR - http://www.scopus.com/inward/record.url?scp=30844464524&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=30844464524&partnerID=8YFLogxK

U2 - 10.1080/03610928608829107

DO - 10.1080/03610928608829107

M3 - Article

AN - SCOPUS:30844464524

SN - 0361-0926

VL - 15

SP - 73

EP - 88

JO - Communications in Statistics - Theory and Methods

JF - Communications in Statistics - Theory and Methods

IS - 1

ER -

Approximating the distribution of the Durbin-Watson statistic

Abstract

All Science Journal Classification (ASJC) codes

Access to Document

Other files and links

Fingerprint

Cite this