The relation of the CCA subspace method to a balanced reduction of an autoregressive model
2004 (English)In: Journal of Econometrics, ISSN 0304-4076, E-ISSN 1872-6895, Vol. 118, no 1-2, 293-312 p.Article in journal (Refereed) Published
In this paper we consider an identification procedure, called MEST, for multivariate time series based on AR-modeling and stochastically balanced truncation and compare it with the CCA subspace method. The stochastically balancing of multivariate AR-models is described using just linear algebraic operations, i.e., no algebraic Riccati equations need to be solved. Both identification procedures are formulated in a uniform manner, and from these expressions we conclude that the only difference is that MEST uses a covariance extension, whereas CCA is based on the sample covariances only. Finally, it is shown that MEST and CCA are asymptotically equivalent.
Place, publisher, year, edition, pages
2004. Vol. 118, no 1-2, 293-312 p.
autoregressive-modeling, maximum entropy covariance extension, stochastically balanced form, canonical correlation analysis, asymptotic analysis
IdentifiersURN: urn:nbn:se:kth:diva-47386DOI: 10.1016/S0304-4076(03)00144-1ISI: 000186669100014ScopusID: 2-s2.0-0346093806OAI: oai:DiVA.org:kth-47386DiVA: diva2:455417
QC 201111102011-11-102011-11-082011-11-10Bibliographically approved