Comparison of R-Forecasting and V-Forecasting Singular Spectrum Analysis in Forecasting Farmers' Exchange Rates in Indonesia

Abstract

Indonesia is an agricultural country where one of the main sources of income comes from the agricultural sector. One of the indicators often used to assess farmer welfare is the Farmer Exchange Rate (FER) index. Research on FER forecasting using the Singular Spectrum Analysis (SSA) method has been widely conducted, however, to date, there has been no research comparing the recurrent forecasting and vector forecasting methods in Indonesia. The purpose of this study is to obtain FER forecasting results using r-forecasting and v-forecasting, then compare the forecasting results based on MAPE values ??to obtain the best forecasting results. The results of the study show that v-forecasting produces better forecasting results with a MAPE value of 0.57%. The forecast results for the next 12 months show an increase and decrease of FER in Indonesia. The highest FER value occurred in May 2022 at 103.79, while the lowest value was in September 2021 at 101.80.