Bangladesh Bank Training Academy Journal

Abstract

This paper concentrates technical analysis to address stochastic and deterministic approach to forecast exchange rate of Bangladesh. Volatility clustering exchange rate of Bangladesh is market driven based on managed fl oat. Vector auto regression (VAR) approach is used to determine the exchange rate in order to form rational expectation regarding exchange rate of Bangladesh. Technical analyses ranging from unit root to VAR have been used to forecast the monthly average exchange rate following new convention. SWAP and forward exchange rate will be deter mined taking into account absolute and comparative advantage and rational ex pectation. I have also applied auto regressive integrated moving average (ARIMA) model to forecast the exchange rate. Non seasonality in ARIMA predicted stable exchange rate for next six months (July-December, 2012), which is mostly close to actual rate. This exercise will help to forming rational expectation about exchange rate forecast from quantitative pointof view.