PERAMALAN NILAI TUKAR PETANI MENGGUNAKAN METODE SINGULAR SPECTRUM ANALYSIS

Abstract

Singular spectrum analysis (SSA) is a new method in time series analysis that uses a nonparametric approach. The purpose of this study is to determine the model and forecast the farmer exchange rate in the Province of Bali using SSA. Vector singular spectrum analysis (VSSA) forecasting method is used to calculate the accuracy of forecasting. The best SSA model is obtained with a window length (L) value of 57 and produces a MAPE value of 0.49%. In conclusion, SSA method can predict farmer exchange rate in the Province of Bali very accurate.