Bootstrapping the Box-Pierce Q test: A robust test of uncorrelatedness

Joel L. Horowitz; I. N. Lobato; John C. Nankervis; N. E. Savin

doi:10.1016/j.jeconom.2005.06.014

Bootstrapping the Box-Pierce Q test: A robust test of uncorrelatedness

Joel L. Horowitz, I. N. Lobato, John C. Nankervis, N. E. Savin^*

^*Corresponding author for this work

Economics

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

50 Scopus citations

Abstract

This paper describes a test of the null hypothesis that the first K autocorrelations of a covariance stationary time series are zero in the presence of statistical dependence. The test is based on the Box-Pierce Q statistic with bootstrap-based P-values. The bootstrap is implemented using a double blocks-of-blocks procedure with prewhitening. The finite sample performance of the bootstrap Q test is investigated by simulation. In our experiments, the performance is satisfactory for samples of n = 500. At this sample size, the differences between the empirical and nominal rejection probabilities are essentially eliminated.

Original language	English (US)
Pages (from-to)	841-862
Number of pages	22
Journal	Journal of Econometrics
Volume	133
Issue number	2
DOIs	https://doi.org/10.1016/j.jeconom.2005.06.014
State	Published - Aug 2006

Keywords

Adjusted P-values
Blocks of blocks bootstrap
Box-Pierce Q
Double bootstrap
Serial correlation tests

ASJC Scopus subject areas

Economics and Econometrics

Access to Document

10.1016/j.jeconom.2005.06.014

Cite this

@article{6de0bc086570452698b064350079837f,

title = "Bootstrapping the Box-Pierce Q test: A robust test of uncorrelatedness",

abstract = "This paper describes a test of the null hypothesis that the first K autocorrelations of a covariance stationary time series are zero in the presence of statistical dependence. The test is based on the Box-Pierce Q statistic with bootstrap-based P-values. The bootstrap is implemented using a double blocks-of-blocks procedure with prewhitening. The finite sample performance of the bootstrap Q test is investigated by simulation. In our experiments, the performance is satisfactory for samples of n = 500. At this sample size, the differences between the empirical and nominal rejection probabilities are essentially eliminated.",

keywords = "Adjusted P-values, Blocks of blocks bootstrap, Box-Pierce Q, Double bootstrap, Serial correlation tests",

author = "Horowitz, {Joel L.} and Lobato, {I. N.} and Nankervis, {John C.} and Savin, {N. E.}",

note = "Funding Information: We thank Paul Weller, the seminar participants at Monash University and participants at the Cowles Foundation Conference “New Developments in Time Series Econometrics” for helpful suggestions and comments. The research of Joel L. Horowitz was supported in part by NSF Grant SES-9910925. I. N. Lobato acknowledges financial support from the Asociaci{\'o}n Mexicana de Cultura and CONACYT Grant no. 41893-S. John Nankervis gratefully acknowledges financial support from the ESRC through Research Grant no. R000222581. ",

year = "2006",

month = aug,

doi = "10.1016/j.jeconom.2005.06.014",

language = "English (US)",

volume = "133",

pages = "841--862",

journal = "Journal of Econometrics",

issn = "0304-4076",

publisher = "Elsevier BV",

number = "2",

}

TY - JOUR

T1 - Bootstrapping the Box-Pierce Q test

T2 - A robust test of uncorrelatedness

AU - Horowitz, Joel L.

AU - Lobato, I. N.

AU - Nankervis, John C.

AU - Savin, N. E.

N1 - Funding Information: We thank Paul Weller, the seminar participants at Monash University and participants at the Cowles Foundation Conference “New Developments in Time Series Econometrics” for helpful suggestions and comments. The research of Joel L. Horowitz was supported in part by NSF Grant SES-9910925. I. N. Lobato acknowledges financial support from the Asociación Mexicana de Cultura and CONACYT Grant no. 41893-S. John Nankervis gratefully acknowledges financial support from the ESRC through Research Grant no. R000222581.

PY - 2006/8

Y1 - 2006/8

N2 - This paper describes a test of the null hypothesis that the first K autocorrelations of a covariance stationary time series are zero in the presence of statistical dependence. The test is based on the Box-Pierce Q statistic with bootstrap-based P-values. The bootstrap is implemented using a double blocks-of-blocks procedure with prewhitening. The finite sample performance of the bootstrap Q test is investigated by simulation. In our experiments, the performance is satisfactory for samples of n = 500. At this sample size, the differences between the empirical and nominal rejection probabilities are essentially eliminated.

AB - This paper describes a test of the null hypothesis that the first K autocorrelations of a covariance stationary time series are zero in the presence of statistical dependence. The test is based on the Box-Pierce Q statistic with bootstrap-based P-values. The bootstrap is implemented using a double blocks-of-blocks procedure with prewhitening. The finite sample performance of the bootstrap Q test is investigated by simulation. In our experiments, the performance is satisfactory for samples of n = 500. At this sample size, the differences between the empirical and nominal rejection probabilities are essentially eliminated.

KW - Adjusted P-values

KW - Blocks of blocks bootstrap

KW - Box-Pierce Q

KW - Double bootstrap

KW - Serial correlation tests

UR - http://www.scopus.com/inward/record.url?scp=33746013821&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=33746013821&partnerID=8YFLogxK

U2 - 10.1016/j.jeconom.2005.06.014

DO - 10.1016/j.jeconom.2005.06.014

M3 - Article

AN - SCOPUS:33746013821

SN - 0304-4076

VL - 133

SP - 841

EP - 862

JO - Journal of Econometrics

JF - Journal of Econometrics

IS - 2

ER -

Bootstrapping the Box-Pierce Q test: A robust test of uncorrelatedness

Abstract

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this