Asymptotic theory for the durbin-watson statistic under long-memory dependence

Shisei Nakamura, Masanobu Taniguchi

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

In time series regression models with "short-memory" residual processes, the Durbin-Watson statistic (DW) has been used for the problem of testing for independence of the residuals. In this paper we elucidate the asymptotics of DW for "long-memory" residual processes. A standardized Durbin-Watson statistic (SDW) is proposed. Then we derive the asymptotic distributions of SDW under both the null and local alternative hypotheses. Based on this result we evaluate the local power of SDW. Numerical studies for DW and SDW are given.

Original languageEnglish
Pages (from-to)847-866
Number of pages20
JournalEconometric Theory
Volume15
Issue number6
Publication statusPublished - 1999
Externally publishedYes

Fingerprint

statistics
Statistics
Asymptotic theory
Long memory
time series
regression

ASJC Scopus subject areas

  • Economics and Econometrics
  • Social Sciences (miscellaneous)

Cite this

Asymptotic theory for the durbin-watson statistic under long-memory dependence. / Nakamura, Shisei; Taniguchi, Masanobu.

In: Econometric Theory, Vol. 15, No. 6, 1999, p. 847-866.

Research output: Contribution to journalArticle

@article{0b6e0cf3f77d4cdb9eab39a2aa28111a,
title = "Asymptotic theory for the durbin-watson statistic under long-memory dependence",
abstract = "In time series regression models with {"}short-memory{"} residual processes, the Durbin-Watson statistic (DW) has been used for the problem of testing for independence of the residuals. In this paper we elucidate the asymptotics of DW for {"}long-memory{"} residual processes. A standardized Durbin-Watson statistic (SDW) is proposed. Then we derive the asymptotic distributions of SDW under both the null and local alternative hypotheses. Based on this result we evaluate the local power of SDW. Numerical studies for DW and SDW are given.",
author = "Shisei Nakamura and Masanobu Taniguchi",
year = "1999",
language = "English",
volume = "15",
pages = "847--866",
journal = "Econometric Theory",
issn = "0266-4666",
publisher = "Cambridge University Press",
number = "6",

}

TY - JOUR

T1 - Asymptotic theory for the durbin-watson statistic under long-memory dependence

AU - Nakamura, Shisei

AU - Taniguchi, Masanobu

PY - 1999

Y1 - 1999

N2 - In time series regression models with "short-memory" residual processes, the Durbin-Watson statistic (DW) has been used for the problem of testing for independence of the residuals. In this paper we elucidate the asymptotics of DW for "long-memory" residual processes. A standardized Durbin-Watson statistic (SDW) is proposed. Then we derive the asymptotic distributions of SDW under both the null and local alternative hypotheses. Based on this result we evaluate the local power of SDW. Numerical studies for DW and SDW are given.

AB - In time series regression models with "short-memory" residual processes, the Durbin-Watson statistic (DW) has been used for the problem of testing for independence of the residuals. In this paper we elucidate the asymptotics of DW for "long-memory" residual processes. A standardized Durbin-Watson statistic (SDW) is proposed. Then we derive the asymptotic distributions of SDW under both the null and local alternative hypotheses. Based on this result we evaluate the local power of SDW. Numerical studies for DW and SDW are given.

UR - http://www.scopus.com/inward/record.url?scp=0033270633&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0033270633&partnerID=8YFLogxK

M3 - Article

AN - SCOPUS:0033270633

VL - 15

SP - 847

EP - 866

JO - Econometric Theory

JF - Econometric Theory

SN - 0266-4666

IS - 6

ER -