FORECASTING EXCHANGE-RATES USING TSMARS

In this article we use the Time Series Multivariate Adaptive Regressio n Splines (TSMARS) methodology to estimate and forecast non-linear str ucture in weekly exchange rates for four major currencies during the 1 980s. The methodology is applied in three steps. First, univariate mod els are fitted to the data and the residuals are checked for outliers. Since significant outliers are spotted in all four currencies, the TS MARS methodology is reapplied in the second step with dummy variables representing the outliers. The empirical residuals of the models obtai ned in the second step pass the standard diagnostic tests for non-line arity, Gaussianity and randomness. Moreover, the estimated models can be sensibly interpreted from an economic standpoint. The out-of-sample forecasts generated by the TSMARS models are compared with those obta ined from a pure random walk. We find that for two of the currencies, the models obtained using TSMARS provide forecasts which are superior to those of a random walk at all forecast horizons. (C) 1998 Elsevier Science Ltd. All rights reserved.