Perbandingan Kinerja Peramalan Kurs di Indonesia

Abstract

Fluctuations in the exchange rate on the money market, both appreciating and depreciating, indicate the volatility that occurs in a country's currency with the currencies of other countries. To overcome the magnitude of the impact of exchange rate fluctuations on the economy, a forecasting model is needed that can predict the exchange rate effectively. This study aims to find the exchange rate forecast that produces the best model in analyzing the exchange rate using the Box-Jenkins/ARIMA, ARCH and GARCH models. The data used in this study is secondary data in a time series pattern in the form of Rupiah/USD exchange rate data obtained from Bank Indonesia in daily form (five days a week), starting from January 2, 2015 to December 31, 2021 with out of sample starting from 3 January 2022 to December 31, 2024. Some of these models are compared with each other so that the best model is obtained, and the forecasting results are 782 days ahead. This study shows that the ARIMA (1,1,0) model is better at predicting the exchange rate than the ARCH (1) model which has the smallest RMSE, MAE and MAPE values. Forecasting results on January 3, 2022 are Rp. 14,298.22/USD with actual data of Rp. 14,270.00/USD. There is shadow forecasting starting from January 3, 2022 to February 11, 2022, so the comparison can be seen with the actual data. For investors, companies or parties with an interest in forecasting the exchange rate, they can use the ARIMA (1,1,0) model in predicting the exchange rate for forecasting several periods in the future.