PREVISÃO DO PREÇO DO AÇÚCAR: UM ESTUDO DE CASO USANDO O SOFTWARE WEKA

Abstract

Abstract: Sugar, since colonization, has always been a prominent product in our country. The forecast of future product prices is very important, in the decision-making process, for producers, market analysts, among others. In this context, this work aims to evaluate the effectiveness of models, provided by WEKA software, for forecasting the price of sugar. The database presents a historical series of monthly sugar prices in the period between January/2006 and July/2020, totaling 175 observations. Multilayer Perception and SMOreg forecast models were used to forecast sugar prices. Results obtained from the two models were compared. It was found, for a short term horizon, that the two forecast models provide reliable estimates for the price of sugar.