Dynamic Hierarchical Factor Models

17 Pages Posted: 15 Dec 2009 Last revised: 31 Jul 2011

See all articles by Emanuel Moench

Emanuel Moench

Deutsche Bundesbank; Goethe University Frankfurt - Department of Money and Macroeconomics

Serena Ng

Columbia Business School - Economics Department

Simon Potter

Peter G. Peterson Institute for International Economics

Date Written: July 26, 2011

Abstract

This paper uses multi-level factor models to characterize within- and between-block variations as well as idiosyncratic noise in large dynamic panels. Block-level shocks are distinguished from genuinely common shocks, and the estimated block-level factors are easy to interpret. The framework achieves dimension reduction and yet explicitly allows for heterogeneity between blocks. The model is estimated using an MCMC algorithm that takes into account the hierarchical structure of the factors. A four-level model is estimated to study block- and aggregate-level dynamics in a panel of 445 series related to different categories of real activity in the United States. The model illustrates the importance of block-level variations in the data.

Keywords: forecasting, monitoring, comovements, large dimensional panel, diffusion index

JEL Classification: C10, C20, C30

Suggested Citation

Moench, Emanuel and Ng, Serena and Potter, Simon, Dynamic Hierarchical Factor Models (July 26, 2011). FRB of New York Staff Report No. 412, Available at SSRN: https://ssrn.com/abstract=1523787 or http://dx.doi.org/10.2139/ssrn.1523787

Emanuel Moench (Contact Author)

Deutsche Bundesbank ( email )

Wilhelm-Epstein-Str. 14
Frankfurt/Main, 60431
Germany
+49 69 95662312 (Phone)

HOME PAGE: http://https://www.bundesbank.de/en/emanuel-moench

Goethe University Frankfurt - Department of Money and Macroeconomics ( email )

Germany

Serena Ng

Columbia Business School - Economics Department ( email )

420 West 118th Street
New York, NY 10027
United States

Simon Potter

Peter G. Peterson Institute for International Economics ( email )

1750 Massachusetts Avenue, NW
Washington, DC 20036
United States

Here is the Coronavirus
related research on SSRN

Paper statistics

Downloads
155
Abstract Views
1,246
rank
209,821
PlumX Metrics