Forecasting Exchange Rate of Bangladesh – A Time Series Econometric Forecasting Modeling Approach

Abstract

In this paper an attempt has been made to select a model for time series forecasting of Average Exchange Rate (AER) of Bangladesh. Our
decision through out this study is mainly concerned with Auto regressive Integrated Moving Average (ARIMA) Model, Holt’s Linear
Exponential Smoothing Model, Simple Linear Regression Model, Log-Linear Regression Model. This project concerns and analyze on a
set of data based on AER during the period July 2003 to June 2007. We try to derive a unique and suitable forecasting model AER. From
our study we find that Holt’s Linear Exponential Smoothing with α=0.999 and β=0.018 gives less forecasting error than that of others. So
we propose that forecasting for the Average Exchange Rate of Bangladesh, one can use the Holt’s Linear Exponential Smoothing Model.
But before using this model one must verify the validation of the model in different time period, because a forecasting model may lose its
validity and suitability as time changes.