Web Application Integrating Time Series Models for Forecasting Corn Production
Cedric M. Caquilala, Perlita E. Gasmen, Alex Gonzaga, Jethro Gem P. Villoria, Loreniel E. Añonuevo, Vince Jebryl Montero, John Riz V. Bagnol · 2024
This study investigates the efficacy of various time series models in forecasting corn production in utilizing localized datasets from a Philippine province. We apply and evaluate models including Seasonal Autoregressive Integrated Moving Average (SARIMA), Bayesian SARIMA, Holt-Winters Exponential Smoothing, and Long Short-Term Memory for its predictive accuracy. Among these models, SARIMA emerges as the top performer, closely followed by Holt-Winters. LSTM and Bayesian SARIMA exhibit varied performances. The study emphasizes the importance of tailored forecasting methods for agricultural decision-making, highlighting SARIMA’s effectiveness in capturing seasonal patterns in corn production. The models are then integrated into a web application for ease of use and presentation of results.