Forecasting economic time series using models of nonstationarity

Van Bellegem, Sébastien [UCL] von Sachs, Rainer [UCL]

The classical forecasting theory of stationary time series exploits the second-order structure (variance, autocovariance and spectral den- sity) of an observed process in order to construct some prediction in- tervals. However, some economic time series show a time-varying un- conditional second-order structure. This article focus on a simple and meaningful model allowing this nonstationary behaviour. We show that this model satisfactory explains the nonstationary behaviour of several economic data sets, among which are the U.S. stock returns and exchange rates. The question how to forecast these processes is addressed and evaluated on the data sets.