期刊名称：Euro Area Balance of Payments and International Investment Position Statistics
印刷版ISSN：1830-3420
电子版ISSN：1830-3439
出版年度：2016
出版社：European Central Bank
摘要：Forecasts from dynamic factor models potentially bene t from re ning the data set by elim-inating uninformative series. The paper proposes to use prediction weights as provided by thefactor model itself for this purpose. Monte Carlo simulations and an empirical application toshort-term forecasts of euro area, German, and French GDP growth from unbalanced monthlydata suggest that both prediction weights and Least Angle Regressions result in improvednowcasts. Overall, prediction weights provide yet more robust results.
关键词：Dynamic factor models; forecasting; variable selection; LARS
