Identification and estimation of dynamic factor models
To reserve a time to meet with the speaker, please register at the following form:
docs.google.com/spreadsheets/d/1y2uYgphPHtmZ2FUYTp1N7DmhER_TiHGciQguJjbJHw4/edit#gid=0

Abstract:
In empirical practice dynamic factor models are typically estimated by performing a Principal Component Analysis (PCA) on the static factor representation. A two-step PC estimator and a sequential least-squares approach is proposed to estimate the original dynamic factors that enter the model with a prespecified number of lags. The identification of the dynamic factors subject to the usual normalization restrictions is analyzed and the consistency of the estimator is established. Monte Carlo simulations suggest that the sequential least-squares estimator outperforms the two-step PCA approach.
Date: 13 October 2017, 14:15 (Friday, 1st week, Michaelmas 2017)
Venue: Manor Road Building, Manor Road OX1 3UQ
Venue Details: Seminar Room C
Speaker: Joerg Breitung (University of Cologne)
Organising department: Department of Economics
Part of: Nuffield Econometrics Seminar
Booking required?: Not required
Audience: Members of the University only
Editors: Erin Saunders, Anne Pouliquen