Modelling and forecasting exchange rates using the Box-Jenkins technique.

Thumbnail Image
Journal Title
Journal ISSN
Volume Title
University of Dar es Salaam
This study aimed at modelling and forecasting monthly exchange rates in Tanzania using the Box Jenkins’s technique. Monthly exchange rates from July 2002 to June 2017 were obtained from Bank of Tanzania. The time series data showed an increasing trend with seasonal fluctuations over time. The time series data were tested for stationarity by Augmented Dickey Fuller test. The data obtained stationarity after the first differencing. Thereafter all procedures for Box Jenkins’s technique were employed, and eventually 12 months forecasts were obtained. SARIMA, SARIMA and SARIMA were the best models for EURO, USD and GBP exchange rates respectively. Selection of the models was based on Akaike’s Information Criterion. Diagnostic check was employed to the fitted models using the Ljung-Box test, autocorrelation function and partial autocorrelation function of the residuals and were found to be normally distributed. Twelve-Month forecasts were obtained from the fitted models and forecasting accuracy measures were calculated as well. The forecasts were seen to be well forecasted indicating the models are adequate. The government, investors and stakeholders in the foreign exchange market can now have a light on the expected future exchange rates and can therefore use the information to formulate proper policies and make decision. Most government projects including infrastructure to be precise are priced in foreign currency, thus having this knowledge the government can have a proper planning of its budget allocation involving foreign currency.
Available in print form, East Africana Collection, Dr. Wilbert Chagula Library, Class mark (THS EAF HG3810.T34N363)
Foreign exchange, Foreign money, The Box-Jenkins technique
Nangay, A. E. (2018). Modelling and forecasting exchange rates using the Box-Jenkins technique. Master dissertation, University of Dar es Salaam. Dar es Salaam.