Yingying Li (Chair)

Abstract

The so-called General Dynamic Factor Model (GDFM), where common shocks are loaded via filters, has many advantages over the more classical "static" Factor Model---where the loadings are matrices. While consistent estimation methods have been developed for the GDFM, no asymptotic distribution results (on the model of Bai, J. (2003) Inferential theory for factor models of large dimensions, Econometrica 71, 135–171) were available so far. Exploiting the duality between common shocks and dynamic loadings, we obtain such results and derive the asymptotic distributions of a class of estimators of the common shocks, the dynamic loadings (filters), the common components, and the impulse response functions. (Joint with Matteo Barigozzi, University of Bologna, Italy, and Paolo Zaffaroni, Imperial College, London)