chapter  3
22 Pages

An Overview of Vector Autoregressive Models

In this chapter, we describe models for a vector time series Xt, with n components X1t,X2t, . . . ,Xnt, observed in times t = 0,±1,±2, . . . . Besides analyzing individual time series Xit, for which the autocorrelation contained in each series is important, we will be interested in dynamic relationships between the component series. We use the notation Xt = (X1t,X2t, . . . ,Xnt)′, t ∈ Z and Xit or Xi,t, for the i-th component, i = 1, . . . , n, where Z is the set of all integers.