Estudo comparativo entre métodos estatísticos e de inteligência artificial para previsão de preço de café no Brasil

Abstract

Forecasting is something common in anyone's daily life, it is a factor of great importance for decision making and strategic planning, even more in a business environment or in the financial market. Thus, there is interest in developing increasingly efficient methodologies to generate these forecasts. This work had as main objective to elaborate a comparative study between statistical and artificial intelligence methods to predict the price of coffee in Brazil. Better performances were obtained with the models LSTM network and ARIMA, however, it was possible to see that each method approached may be suitable in relation to some means of use. Thus, at different scales, the models discussed can be useful for decision making.