On forecast of exchange rate of a foreign currency

Summarizes research on methods for forecasting the exchange rate of Italian Lire, British Pounds and German Marks in US Dollars. Different calculus procedures and forecast quality criteria were analyzed based on numerical data for the last two decades. Several methods, including time-series methods and a common correlation method, were developed to forecast exchange rates. Two quality criteria, future price criterion (FPC) and risk charge criterion (RCC), were formulated to measure the accuracy of the forecasting methods. The analysis showed that the FPC criterion alone could not effectively determine the best method because all optimal values were very close. The second criterion, RCC, was more complicated but gave more information about the methods and the data. Based on the RCC criterion, a better forecast was achieved using the optimal length of identification interval. The associated risk premium charge was also obtained. The common correlation method was found to be the best method with the identification interval length equal to 72-months for all currencies being considered. This method can be used as an analytical tool in estimating the risk involved and associated premium charges when negotiating a forward contract of foreign currencies.