DP2400 Reference Cycles: The NBER Methodology Revisited

Author(s): Mario Forni, Marc Hallin, Marco Lippi, Lucrezia Reichlin
Publication Date: March 2000
Keyword(s): Coincident And Leading Indicators, Dynamic Factor Models, Dynamic Principal Components Series
JEL(s): C13, C33, C43
Programme Areas: International Macroeconomics
Link to this Page: cepr.org/active/publications/discussion_papers/dp.php?dpno=2400

This paper proposes a new way to compute a coincident and a leading index of economic activity. The method provides a unified approach for the selection of the coincident and the leading variables, for averaging them into coincident and leading indexes and for the identification of turning points. The statistical framework we propose reconciles dynamic principal components analysis wit dynamic factor analysis. We use our procedure to estimate coincident and leading indexes for the EMU area as well as country-specific indexes. Unlike other methods used in the literature, the country indexes take into consideration the cross-country as well as the within-country correlation structure.