Application of teh Hybrid Singular Spectrum Analysis – ARIMA Model for Indonesia's Inflation Rate (2018-2023)

Sri Rahayu (Universitas Negeri Makassar)
Aswi Aswi (Universitas Negeri Makassar)
Muhammad Fahmuddin Sudding (Universitas Negeri Makassar)

Jurnal Matematika Sains dan Teknologi
Vol. 25 No. 2 (2024): September (in Progress)

This research aims to determine the results and accuracy of forecasting inflation rates in Indonesia using Hybrid Singular Spectrum Analysis (SSA) – Autoregressive Integrated Moving Average (ARIMA). Hybrid SSA-ARIMA combines two time series methods to increase forecasting accuracy, especially for economic data that contains trend and seasonal components. The data used is data on the national consumer price inflation rate (Y-on-Y) for the period January 2018 to December 2023. The forecast accuracy obtained by the MAPE value for Singular Spectrum Analysis was 56.26797%, and Hybrid SSA-ARIMA was 18.88851%. This shows that Hybrid SSA-ARIMA has better forecasting capabilities than Singular Spectrum Analysis in predicting the inflation rate in Indonesia.