1468-0084/asset/OBES_left.gif?v=1&s=5448fd051d737557110d1d043286851c7838fad2)
1468-0084/asset/OBES_right.gif?v=1&s=971810f34166a1204f758a72f2fb3ed4b5fb38a6)
A Canonical Correlation Approach for Selecting the Number of Dynamic Factors
Article first published online: 6 NOV 2012
DOI: 10.1111/obes.12003
© Blackwell Publishing Ltd and the Department of Economics, University of Oxford 2012
Issue
Oxford Bulletin of Economics and Statistics
Special Issue: Large Data Sets
Volume 75, Issue 1, pages 23–36, February 2013
Additional Information
How to Cite
Breitung, J. and Pigorsch, U. (2013), A Canonical Correlation Approach for Selecting the Number of Dynamic Factors. Oxford Bulletin of Economics and Statistics, 75: 23–36. doi: 10.1111/obes.12003
Publication History
- Issue published online: 21 DEC 2012
- Article first published online: 6 NOV 2012
- Final manuscript Received: September 2012
- Abstract
- Article
- References
- Cited By
Keywords:
- C33;
- C52
Abstract
In this article, we propose a selection procedure that allows us to consistently estimate the number of dynamic factors in a dynamic factor model. The procedure is based on a canonical correlation analysis of the static factors which has the advantage of being invariant to a rescaling of the factors. Monte Carlo simulations suggest that the proposed selection rule outperforms existing ones, in particular, if the contribution of the common factors to the overall variance is moderate or low. The new selection procedure is applied to the US macroeconomic data panel used in Stock and Watson [NBER working paper 11467 (2005)].