A Vector Autoregressive Forecasting Model of The US$/$A Exchange Rate

Abstract
A forecasting model of the US$/$A exchange rate is derived through the application of vector autoregression (VAR) techniques. The major theoretical models of exchange rate determination are reviewed to identify relevant variables to include in the VAR model. For the within-sample period of September 1974 to May 1983, the VAR forecasts are found to be clearly superior to the naive no-charge extrapolated forecasts. However, the position is reversed when the out-of-sample forecasts are examined.

This publication has 10 references indexed in Scilit: