Heterogeneidade sazonal do comércio exterior brasileirouma análise regional usando o método de saturação de indicadores
Silva Junior, Luiz Fernando Lopes da
O documento é disponibilizado pela fonte de origem, que mantém a versão integral e as condições de uso.
Resumo
This study analyzes the seasonal heterogeneity of Brazilian foreign trade at the regional level (North, Northeast, Center-West, Southeast, and South) from 2000 to 2024. Using the econometric approach of Indicator Saturation (ISAT) combined with the General-to-Specific (GETS) modeling framework on monthly real volume series, the research investigates the presence of seasonality, calendar effects, and structural shocks in export and import flows. The results indicate that seasonality is predominantly stochastic, captured by an autoregressive (AR) structure, while the number of working days proved to be the only significant calendar effect across all regions. The ISAT methodology identified structural breaks and shocks associated with known economic events, such as customs strikes (2002), the commodity supercycle (2003-2008), the global financial crisis (2008-2009), and the severe drought in the North region (2023-2024). Distinct regional dynamics were observed, with the Central-West region leading the average growth in exports, and the Northeast leading in imports. The study concludes that aggregate national analysis masks important regional patterns and demonstrates the robustness of the GETS/ISAT method for modeling time series subject to shocks, offering insights for trade policy and logistical planning.
Ficha do documento
- Tipo
- Dissertação
- Ano
- 2025
- Instituição
- Fundação Getulio Vargas
- Fonte
- Repositório da FGV
- Idioma
- Português
- Acesso
- Acesso aberto
- Identificador
- oai:repositorio.fgv.br:10438/37556
Conteúdos relacionados
- TeseEnsaios sobre previsão de vendas no varejo de modaFundação Getulio Vargas · 2021
- DissertaçãoRegimes' characteristics and time series forecastingFundação Getulio Vargas · 2026
- DissertaçãoNowcasting GDPFundação Getulio Vargas · 2026
- DissertaçãoModelo preditivo para precificação de ativos com integração de notícias do mercado financeiro e técnicas de machine learningFundação Getulio Vargas · 2025