Comparison Between ARIMA and VAR Model Regarding the Forecasting of the Price of Jute Goods in Bangladesh

Abstract

In this study we used Autoregrressive Intigrated Moving Average (ARIMA) and Vector Autoregrressive (VAR) model to analyze and forecast the price of total Jute Goods with four of its types, where data has been collected from Bangladesh Jute Mills Corporation (BJMC) from the year 1980-81 to 2013-2014. In this study, a comparison has been made regarding ARIMA model and VAR model to investigate which model is the best to forecast. The methodology employed in this study is the co-integration and Granger Causality under VECM. The Augmented Dickey Fuller (ADF) Test has been performed to test the stationarity of the data set. The findings of this study suggested that in forecasting the price of jute goods of Bangladesh, the ARIMA model is more efficient than VAR model.