Peramalan Total Nilai Ekspor Indonesia Menggunakan Metode Singular Spectrum Analisis

Abstrak

This research aims to forecast the total export value of two main sectors in Indonesia for the period of January to December 2024 using the Singular Spectrum Analysis (SSA) method. Forecasting plays a crucial role in supporting economic policy planning and strategic decision-making, particularly in the export sector, which significantly contributes to national income. The SSA method is selected due to its ability to decompose time series data into interpretable components such as trend, seasonal, and noise. The performance of the forecasting model is evaluated using the Mean Absolute Percentage Error (MAPE), which offers an intuitive interpretation in percentage form. Based on the results, the optimal parameter in SSA is obtained at L=28, with a MAPE value of 16.63%, indicating good forecasting accuracy. The forecasted export value shows that the highest export is expected in December 2024, reaching 39.578,67 million USD, while the lowest is projected in January 2024 at 21.689,14 million USD. These findings demonstrate that SSA is effective in forecasting economic time series data, particularly Indonesia’s export values. This research not only contributes to the practical application of SSA in economics but also serves as a reference for future studies and policymakers in formulating export strategies.