Modelo de previsão de mercado para o consumo de óleo diesel no Brasil

Loading...
Thumbnail Image

Journal Title

Journal ISSN

Volume Title

Publisher

Universidade Federal de Juiz de Fora (UFJF)

DOI

Abstract

The main objective of this work is predict and analyze the consumption of diesel in Brazil. The results obtained in the study can assist the economic agents in relation to the future demand, since it is an important fuel and used in several areas of the country. The database consists of monthly data of the daily average of diesel fuel consumption, measured in thousands of barrels, between the periods of January 1979 and January 2023. This information represents the apparent consumption corresponding to domestic industrial production increased imports and decreased exports. The methods used were the exponential smoothing (ES) and Box & Jenkins (BJ) method. It was made a recursive analysis out of the sample to choose the best method by performing a competition between the two. The winning method was BJ, at the end of the competition analysis, an ARIMA(1,1,1)(1,0,2) with square root transformation. This was the model used to calculate diesel oil consumption forecasts in Brazil until December 2024. Through the estimated values it was found that it is possible to increase consumption in 2023 and 2024.

Description

Citation

Endorsement

Review

Supplemented By

Referenced By

Creative Commons license

Except where otherwise noted, this item's license is described as Acesso Aberto