Optimization of Average Rice Price Forecasting in East Java Using Decomposition-Weighted Fuzzy Time Series with Lagrange and Differential Evolution

Agus Fachrur Rozy (Department of Statistics, University Brawijaya, Malang)
Alya Fitri Syalsabilla (Department of Statistics, Institut Teknologi Sepuluh Nopember, Surabaya)

Jurnal Ekonomi dan Pembangunan
Vol. 32 No. 2 (2024): In Progress

Accurate rice price forecasting is crucial for East Java's economic stability. East Java stands as Indonesia's top rice-producing province, supported by data from Badan Pusat Statistik (BPS) published in October 2024, which highlights an impressive production volume of 9 million tons, outperforming all other provinces in the country. Using the Weighted Fuzzy Time Series (WFTS) model combined with Lagrange Quadratic Programming (LQP) and Differential Evolution (DE), this study forecasts rice prices from January to July 2024 using data from https://panelharga.badanpangan.go.id. The WFTS-LQP-DE model enhances accuracy by refining forecasts through DE, showing improvements over the WFTS-LQP model with lower Mean Squared Error (MSE), Root Mean Squared Error (RMSE), and Mean Absolute Error (MAE). Recommendations include integrating WFTS-LQP-DE forecasts into planning, improving supply chain management, and establishing a price stabilisation fund.