Nur Alvi Hasanah, Trianingsih Eni Lestari
Forecasting is a process to predict future events based on past data. In this article, forecasting of Farmers' Terms of Trade (FTT) in 5 provinces of Java, namely West Java, Central Java, Yogyakarta, East Java, and Banten, uses the Multivariate Singular Spectrum Analysis (MSSA) method. The purpose of Forecasting Farmers' Terms of Trade (FTT) is to see the level of welfare of farmers who play a role in national income. The MSSA method has the advantage of excellent forecasting accuracy and can be applied to data that has a linear or nonlinear relationship. The results obtained are the grouping of 4 groups based on 8 eigenvectors, with the resulting forecast length of 13 periods. The level of accuracy of forecasting data is obtained from the Mean Absolute Percentage Error (MAPE) in each province with the highest MAPE of 0.71% occurring in East Java and the lowest being 0.33% in Yogyakarta. Based on the MAPE results, it is less than 10% so that forecasting of Farmers' Terms of Trade (FTT) for the period April 2021 to April 2022 using the Multivariate Singular Spectrum Analysis method is included in the Forecasting category with very good ability. The forecasting results show that there is a seasonal pattern in each variable for the next 13 periods. © 2022 American Institute of Physics Inc.. All rights reserved.
Mathematics Department, Faculty of Mathematics and Natural Sciences, Universitas Negeri Malang, Malang, Indonesia