Forecasting the Export Unit Value Index in Indonesia Using the Single Input Transfer Function

Abstract

Export is one of the factors that increase the economic growth of a country. One measure of export activity that can describe economic growth in Indonesia is The Export Unit Value Index, which is an index that measures changes in the price of export commodities sold by residents of one country to residents of other countries. The purpose of this study is to predict the unit value index of exports in Indonesia using a single input transfer function model and to see the influence of the value of oil and gas and non-oil and gas exports on the unit value index of exports in Indonesia. The single input transfer function model is a model that describes the future forecast of a series (output series) obtained based on the past values of the output series and other time series (input series) that affect the output series. The results of this study obtained a transfer function model with the order (0,0,1) with a noise series following ARIMA (1,0,1). Based on this model, the export unit value index at time t is influenced by the unit value export index in the previous month and is influenced by the oil and gas and non-oil and gas export value in the same month. As indicated by its MAPE value of 4.89%, the forecast value does not diverge much from the actual value, which suggests that the transfer function model can be used to predict the export unit value index in Indonesia.
 Keywords: forecasting, export unit value index, single input, transfer function