TY - JOUR
T1 - Selecting models with different spectral density matrix structures by the cross-validated log likelihood criterion
AU - Matsuda, Yasumasa
AU - Yajima, Yoshihiro
AU - Tong, Howell
N1 - Copyright:
Copyright 2009 Elsevier B.V., All rights reserved.
PY - 2006/4
Y1 - 2006/4
N2 - We propose the cross-validated log likelihood (CVLL) criterion for selecting multivariate time series models with different forms of the spectral density matrix, which correspond to different constraints on the component time series such as mutual independence, separable correlation, time reversibility, graphical interaction and others. We obtain asymptotic properties of the CVLL, and demonstrate the empirical properties of the CVLL selection with both simulated and real data.
AB - We propose the cross-validated log likelihood (CVLL) criterion for selecting multivariate time series models with different forms of the spectral density matrix, which correspond to different constraints on the component time series such as mutual independence, separable correlation, time reversibility, graphical interaction and others. We obtain asymptotic properties of the CVLL, and demonstrate the empirical properties of the CVLL selection with both simulated and real data.
KW - Conditional independence
KW - Consistency
KW - Graphical model
KW - Kullback-leibler divergence
KW - Model selection
KW - Multivariate time series
KW - Periodogram
KW - Spectral density matrix
UR - http://www.scopus.com/inward/record.url?scp=33745622264&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=33745622264&partnerID=8YFLogxK
U2 - 10.3150/bj/1145993973
DO - 10.3150/bj/1145993973
M3 - Article
AN - SCOPUS:33745622264
VL - 12
SP - 221
EP - 249
JO - Bernoulli
JF - Bernoulli
SN - 1350-7265
IS - 2
ER -