Abstract The paper describes a new, fully recursive method for identifying, estimat-ing and forecasting multivariate (vector) time-series. Any low frequency (trend) components associated with each of the elements of the vector time-series are first removed by recursive, fixed interval smoothing based on generalised random walk (GRW) models; while the vector of perturbational residuals obtained from this “detrending” step is then modelled as a vector AR or ARMA process. Finally the various structural models are combined to yield an overall multivariate, state-space model, which provides the basis for forecasting, using standard Kalman Filter methods. The practical utility of the approach is illustrated by a sales forecasting example.
[1]
Peter C. Young,et al.
Recursive Estimation, Forecasting, and Adaptive Control
,
1989
.
[2]
G. Kitagawa,et al.
A Smoothness Priors–State Space Modeling of Time Series with Trend and Seasonality
,
1984
.
[3]
Andrew Harvey,et al.
A unified view of statistical forecasting procedures
,
1984
.
[4]
Peter C. Young,et al.
The Instrumental Variable Method: A Practical Approach to Identification and System Parameter Estimation
,
1985
.
[5]
R. K. Mehra,et al.
Identification in Control and Econometrics1)
,
1982
.