Data: 09/02/2010 - 12.30 – 14.30

Autore: Cecilia Frale - Libero Monteforte - MEF – Banca d’Italia

Abstract: The literature on short term forecasting suggests that there is room for improvement in forecast ability considering a richer dynamic structure. To address this issue, multivariate factor models are combined with MIDAS dynamics to perform short-term forecasts. A dynamic factor model with mixed frequency is proposed, where the past observations of high frequency indicators are included following the MIDAS approach. This structure is able to represent in a richer dynamics the information content of the economic indicators and produces smoothed factors and forecasts. The short term forecasting performance of the model is evaluated against other models in a pseudo-real time experiment, also allowing for pooled forecast from factor models.
Keywords: Temporal Disaggregation. Dynamic factorModels. MIDAS. Forecasting.
J.E.L. Classification: E32, E37, C53