Modelo de previsisão de mercado para o consumo de gasolina no Brasil
Loading...
Date
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Universidade Federal de Juiz de Fora (UFJF)
DOI
Abstract
The main objective of this work is to forecast and analyze gasoline consumption in
Brazil. Due to the significant influence of this fuel on the Brazilian energy matrix, the
importance of a study focused on data projection for this derivative is justified.
Consequently, the information provided can serve as a basis for economic agents
decisions. The database consists of monthly data on the daily average gasoline
consumption, measured in thousands of barrels, from January 1979 to March 2024.
The selected data highlight the apparent consumption equivalent to domestic industrial
production plus imports and minus exports. The applied methods were the Exponential
Smoothing Method (ESM) and the Box & Jenkins Method (BJM). An out-of-sample
recursive analysis was conducted to select the most effective method, resulting in a
competition between them. The winning method (ESM), using a Holt-Winters model,
provided the gasoline consumption forecasts in Brazil until December 2025. The
results indicate a possible decline in consumption in 2025.
Description
Citation
Endorsement
Review
Supplemented By
Referenced By
Creative Commons license
Except where otherwise noted, this item's license is described as Acesso Aberto

