Statistical Modelling 10 (2010), 353–374

The vector innovations structural time series framework: a simple approach to multivariate forecasting

Ashton de Silva
School of Economics, Finance and Marketing,
RMIT University,
VIC 300
Australia
eMail: ashton.desilva@rmit.edu.au

Rob J Hyndman and Ralph Snyder
Department of Econometrics and Business Statistics,
Monash University
Australia

Abstract:

The vector innovations structural time series framework is proposed as a way of modelling a set of related time series. As with all multivariate approaches, the aim is to exploit potential interseries dependencies to improve the fit and forecasts. The model is based around an unobserved vector of components representing features such as the level and slope of each time series. Equations that describe the evolution of these components through time are used to represent the inter-temporal dependencies. The approach is illustrated on a bivariate dataset comprising Australian exchange rates of the UK pound and US dollar. The forecasting accuracy of the new modelling framework is compared to other common uni- and multivariate approaches in an experiment using time series from a large macroeconomic database.

Keywords:

exponential smoothing; forecast comparison; multivariate time series; state space model;vector autoregression; vector innovations structural time series model
back