摘要
This work introduces a methodology to estimate coffee prices based on the use of Extreme Learning Machines.The process is initiated by identifying the presence of nonstationary components,like seasonality and trend.These components are withdrawn if they are found.Next,the temporal lags are selected based on the response of the Partial Autocorre-lation Function filter.As predictors,we address the following models:Exponential Smooth-ing(ES),Autoregressive(AR)and Autoregressive Integrated and Moving Average(ARIMA)models,Multilayer Perceptron(MLP)and Extreme Learning Machines(ELMs)neural net-works.The computational results based on three error metrics and two coffee types(Ara-bica and Robusta)showed that the neural networks,especially the ELM,can reach higher performance levels than the other models.The methodology,which presents preprocess-ing stages,lag selection,and use of ELM,is a novelty that contributes to the coffee prices forecasting field.
基金
The authors would like to thank the National Council for Sci-entific and Technological Development(CNPq-Brazil),processes number 405580/2018-5 and 315298/2020-0
the Arauca´ria Foundation,process number 51497,for their finan-cial support.
