Testing for Zero Autocorrelation in the Presence of Statistical Dependence

I. N. Lobato; John C. Nankervis; N. E. Savin

doi:10.1017/S0266466602183083

Back

Testing for Zero Autocorrelation in the Presence of Statistical Dependence

Journal article

Peer reviewed

Testing for Zero Autocorrelation in the Presence of Statistical Dependence

I. N. Lobato, John C. Nankervis and N. E. Savin

Econometric theory, Vol.18(3), pp.730-743

06/01/2002

DOI: 10.1017/S0266466602183083

View Online

Abstract

The problem addressed in this paper is to test the null hypothesis that a time series process is uncorrelated up to lag K in the presence of statistical dependence. We propose an extension of the Box-Pierce Q-test that is asymptotically distributed as chi-square when the null is true for a very general class of dependent processes that includes non-martingale difference sequences. The test is based on a consistent estimator of the asymptotic covariance matrix of the sample autocorrelations under the null. The finite sample performance of this extension is investigated in a Monte Carlo study.

Autocorrelation

Consistent estimators

Covariance

Covariance matrices

Critical values

Estimators

Null hypothesis

Sample size

Time series

Time series models

Details

Title: Subtitle: Testing for Zero Autocorrelation in the Presence of Statistical Dependence
Creators: I. N. Lobato - Instituto Tecnológico Autónomo de México
John C. Nankervis - University of Surrey
N. E. Savin - University of Iowa
Resource Type: Journal article
Publication Details: Econometric theory, Vol.18(3), pp.730-743
DOI: 10.1017/S0266466602183083
ISSN: 0266-4666
eISSN: 1469-4360
Publisher: Cambridge University Press
Number of pages: 14
Language: English
Date published: 06/01/2002
Academic Unit: Economics
Record Identifier: 9984963201802771

Metrics

1 Record Views