A two-step estimator for large approximate dynamic factor models based on Kalman filtering

Catherine Doz (1, 2) , Domenico Giannone (3) , Lucrezia Reichlin

1 PSE - Paris School of Economics
2 CES - Centre d'économie de la Sorbonne
3 ECARES - European Center for Advanced Research in Economics and Statistics

This paper shows consistency of a two step estimation of the factors in a dynamic approximate factor model when the panel of time series is large (n large). In the first step, the parameters of the model are estimated from an OLS on principal components. In the second step, the factors are estimated via the Kalman smoother. The analysis develops the theory for the estimator considered in Giannone, Reichlin, and Sala (2004) and Giannone, Reichlin, and Small (2008) and for the many empirical papers using this framework for nowcasting.

Mots clés: Factor Models; Kalman Filter; Principal Components; Large Cross-sections

Domaines: Economies et finances; Méthodes et statistiques

Catherine Doz, Domenico Giannone, Lucrezia Reichlin. A two-step estimator for large approximate dynamic factor models based on Kalman filtering. Econometrics, 2011, 164 (1), pp.188-205. ⟨hal-00638009⟩