| Nome: | Descrição: | Tamanho: | Formato: | |
|---|---|---|---|---|
| 716.39 KB | Adobe PDF |
Orientador(es)
Resumo(s)
This Dissertation focuses on exploring Dynamic Factor Models methodologies allowing for an
e ccient and reliable forecast of Quarterly real GDP growth rates on a timely maner. Alternative
forecasting models will be described, together with the most recent litterature focusing on the ex-
perience of the European Monetary Union (EMU), Spain and Portugal. The accuracy of the created
Dynamic Factor Model will be compared with the Portuguese Statistical O ce preliminary announce-
ment, 45 days after the end of the reference quarter. The nature of the economic variables to be used,
their timeliness and the weight given by the model to their information content are analyzed, in the
search for undoing the black box character of factor models. A wide range of future developments
are identi ed while acquiring the understanding that a Forecaster will always have to deal with open
questions.