The Generalised Dynamic Factor Model: One Sided Estimation and Forecasting
Journal of the American Statistical Association, 2005.
Reichlin, L., Forni, M., Hallin, M., Lippi, M.
Date Published: 01/09/2005
Abstract:
This paper proposes a new forecasting method which makes use of information from a large panel of time series. As in Forni, Hallin, Lippi and Reichlin (2000), and in Stock and Watson (2002a,b), the method is based on a dynamic factor model. We argue that our method improves upon a standard principal component predictor in that, first, it fully exploits all the dynamic covariance structure of the panel and, second, it weights the variables according to their estimated signal-to-noise ratio. We provide asymptotic results for our optimal forecast estimator and show that in finite samples our forecast outperforms the standard principal components predictor.
Citation:
Reichlin, L., Forni, M., Hallin, M., Lippi, M., "The Generalised Dynamic Factor Model: One Sided Estimation and Forecasting", Journal of the American Statistical Association, Vol. 100, No. 471, September 2005 pp.830-840 (11)
Search Google Scholar for this entry