Perbandingan Fuzzy Time Series Lee dan Double Exponential Smoothing pada Permalan Garis Kemiskinan Provinsi Gorontalo

Abstract

This study aims to compare the forecasting accuracy of two statistical methods, namely Fuzzy Time Series (FTS) Model Lee and Double Exponential Smoothing (DES) Holt, in predicting the poverty line in Gorontalo Province. The data used were secondary data obtained from the Central Bureau of Statistics (BPS) of Gorontalo Province for the period 2004–2024. A quantitative descriptive approach with time series analysis was applied. The analysis process involved constructing the universe of discourse, fuzzification, and defuzzification for the FTS method, as well as double smoothing with parameters α and β for the DES method. The findings revealed that the FTS Lee model achieved the highest forecasting accuracy with a Mean Absolute Percentage Error (MAPE) of 0.47%, while the DES Holt method yielded a MAPE of 2.26%. The forecasting results indicated that Gorontalo’s poverty line is projected to reach IDR 464,959.50 in 2025 and IDR 504,911.90 in 2026. These results suggest that the FTS Lee method is more effective in predicting socio-economic data characterized by high uncertainty.