A UNIFIED APPROACH TO IDENTIFYING MULTIVARIATE TIME-SERIES MODELS

This article proposes a Bayesian procedure for simultaneous identifica tion of the Kronecker indices and model parameters of a multivariate l inear system. The model parameters include the starting values and inn ovations of the system so that the series considered may be co-integra ted or non-invertible. The procedure uses some recent developments in stochastic search variable selection in linear regression analysis and Markov chain Monte Carlo methods in statistical computing. It also ta kes into consideration the row structure of a vector model implied by the Kronecker indices. Comparison with other existing methods is discu ssed. Simulated and real examples are used to illustrate the proposed procedure.