TY - JOUR

T1 - On the distribution of the residual cross-correlations of infinite order vector autoregressive series and applications

AU - Bouhaddioui, Chafik

AU - Roy, Roch

N1 - Funding Information:
This work was supported by grants to the second named author from the Natural Science and Engineering Research Council of Canada, the Network of Centers of Excellence on the Mathematics of Information Technology and Complex Systems (MITACS) and the Fonds québécois de la recherche sur la nature et les technologies (FQRNT).

PY - 2006/1/1

Y1 - 2006/1/1

N2 - Here we derive the asymptotic distribution of an arbitrary vector of residual cross-correlations resulting from the fitting of finite autoregressions to two uncorrelated infinite order vector autoregressive series. Its asymptotic distribution is the same multivariate normal as the one of the corresponding vector of cross-correlations between the two innovation series. The application of that result for testing the uncorrelatedness of two series is also discussed.

AB - Here we derive the asymptotic distribution of an arbitrary vector of residual cross-correlations resulting from the fitting of finite autoregressions to two uncorrelated infinite order vector autoregressive series. Its asymptotic distribution is the same multivariate normal as the one of the corresponding vector of cross-correlations between the two innovation series. The application of that result for testing the uncorrelatedness of two series is also discussed.

KW - Asymptotic distribution

KW - Finite autoregression

KW - Portmanteau statistics

KW - Residual cross-correlations

KW - Tests for non-correlation

UR - http://www.scopus.com/inward/record.url?scp=30944452979&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=30944452979&partnerID=8YFLogxK

U2 - 10.1016/j.spl.2005.07.004

DO - 10.1016/j.spl.2005.07.004

M3 - Article

AN - SCOPUS:30944452979

SN - 0167-7152

VL - 76

SP - 58

EP - 68

JO - Statistics and Probability Letters

JF - Statistics and Probability Letters

IS - 1

ER -