A carregar...
Publicação
Forecasting Portuguese GDP growth using U-MIDAS and VAR models: a comparative analysis
| Nome: | Descrição: | Tamanho: | Formato: | |
|---|---|---|---|---|
| 457.41 KB | Adobe PDF |
Orientador(es)
Resumo(s)
This study examines the benefits of using high-frequency data in forecasting Portuguese GDP
growth, comparing U-MIDAS and VAR models. U-MIDAS integrates monthly data for
flexibility, while VAR uses quarterly data. Using data from 2000Q1 to 2020Q4 for model
estimation, forecasts cover 2021Q1 to 2024Q1, a period of COVID-19 recovery and
inflationary pressures. Key indicators include industrial production, unemployment rate, PSI 20 index, and exports. The results show that U-MIDAS performs better in short- and long-term
forecasts, while VAR excels in the medium term. These findings are statistically compared
using methods such as the Diebold-Mariano test, and the Clark-West test.
Descrição
Palavras-chave
Forecasting Mixed data sampling U-MIDAS VAR Portuguese GDP Nowcasting High-frequency data