Variable step-size evolving participatory learning with kernel recursive least squares applied to gas prices forecasting in Brazil

Eduardo Ravaglia Campos Queiroz; Kaike Sa Teles Rocha Alves; Fernando Luiz Cyrino Oliveira; Eduardo Pestana de Aguiar

doi:10.1007/s12530-021-09388-z

Variable step-size evolving participatory learning with kernel recursive least squares applied to gas prices forecasting in Brazil

Evol Syst (Berl). 2022;13(2):297-306. doi: 10.1007/s12530-021-09388-z. Epub 2021 May 25.

Authors

Eduardo Ravaglia Campos Queiroz¹, Kaike Sa Teles Rocha Alves², Fernando Luiz Cyrino Oliveira¹, Eduardo Pestana de Aguiar³

Affiliations

¹ Department of Industrial Engineering, Pontifical Catholic University of Rio de Janeiro, Rio de Janeiro, RJ Brazil.
² Graduate Program in Computational Modeling, Federal University of Juiz de Fora, Juiz de Fora, MG Brazil.
³ Department of Industrial and Mechanical Engineering, Federal University of Juiz de Fora, Juiz de Fora, MG Brazil.

Abstract

A prediction model is an indispensable tool in business, helping to make decisions, whether in the short, medium, or long term. In this context, the implementation of machine learning techniques in time series forecasting models has a notorious relevance, as information processing and efficient and dynamic knowledge uncovering are increasingly demanded. This paper develops a model called Variable step-size evolving Participatory Learning with Kernel Recursive Least Squares, VS-ePL-KRLS, applied to the forecast of weekly prices for S500 and S10 diesel oil, at the Brazilian level, for biweekly and monthly horizons. The presented model demonstrates a better accuracy compared with analogous models in the literature, without loss of computational performance for all time series analyzed.

Keywords: Evolving fuzzy models; Forecasting; Time series; Variable step-size.