Damásio, Bruno Miguel PintoPinheiro, Flávio Luís PortasBorges, Bruno Roovers De Avelar Esteves2023-02-242023-02-242023-01-25http://hdl.handle.net/10362/149643Dissertation presented as the partial requirement for obtaining a Master's degree in Statistics and Information Management, specialization in Information Analysis and ManagementThe measuring of economic activity has long been a time consuming process which often results in official aggregated statistics being published with some delay in relation to the end of the observation period. In order to assess where the forecasting of country wide economic activity can be the most useful, we propose an exercise on which the forecasts root mean squared errors of several European countries’ GDP are taken trough a MIDAS regression on a rolling estimation window. The results of this exercise should allow for a comparison on the regional dispositions of the forecasting quality and assess which countries show the highest gains from the use of MIDAS methodology in comparason to the ARIMA univariate process. We find that the gains of using the MIDAS methodology in relation to the ARIMA process are heterogenous between countries. Even though the countries appear to have distinct potential for the use of MIDAS models, the geographical sub-region division into Center, Eastern, Northern and Southern European countries does not seem to produce good separation groups.engMIDASForecasting QualityEconomic ActivityEuropean Coutriesmaster thesis203237170