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

• Published in: Econometrics 2011-09, Vol.164 (1), p.188-205
• Main Authors: Doz, Catherine, Giannone, Domenico, Reichlin, Lucrezia
• Format: Article
• Language: English
• Subjects: Applications; Distribution theory; Dynamic models; Econometric models; Econometrics; Estimating techniques; Estimation; Exact sciences and technology; Factor analysis; Factor models; Factor models Kalman filter Principal components Large cross-sections; Inference from stochastic processes; time series analysis; Insurance, economics, finance; Kalman filter; Kalman filters; Large cross-sections; Mathematics; Model testing; Multivariate analysis; Panel data; Principal components; Principal components analysis; Probability and statistics; Probability theory; Quantitative Finance; Regression analysis; Sciences and techniques of general use; Significance tests; Statistical Finance; Statistical models; Statistics; Studies; Time series
• ISSN: 0304-4076; 2225-1146; 1872-6895; 2225-1146
• DOI: 10.1016/j.jeconom.2011.02.012

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 et al. (2004) and Giannone et al. (2008) and for the many empirical papers using this framework for nowcasting.