Abstract
This paper compares the forecasting performance of three alternative factor models based on business survey data for the industrial production in Italy. The first model uses static principal component analysis, while the other two apply dynamic principal component analysis in frequency domain and subspace algorithms for state-space representation, respectively. Once the factors are extracted from the business survey data, then they are included into a single equation to predict the industrial production index. The forecast results show that the three factor models have a better performance than that of a simple autoregressive benchmark model regardless of the specification and estimation methods. Furthermore, the state-space model yields superior forecasts amongst the factor models.
Acknowledgements
The author thanks two anonymous referees for their valuable suggestions that have improved the first version of this work. The author also thanks Patrizia Margani, Carmine Pappalardo and Claudio Lupi for their helpful comments.
Notes
The models have been used in other studies not related to Italy (see, e.g. Citation27).
The data were collected before many of the scientific and institutional functions carried out by Italian Institute for Studies and Economic Analyses (ISAE) and were transferred to ISTAT.
The business survey data are not subject to revisions while this happens to the industrial production index. In this respect, Bulligan et al. Citation8 show that using indicators not subject to revisions suggest the virtual irrelevance of the real-time issue for the forecasting ability.
I would like to thank a referee for having drawn my attention on the issue regarding the use of different IC.