Fosten, Jack (2017) Model selection with estimated factors and idiosyncratic components. Journal of Applied Econometrics, 32 (6). 1087–1106. ISSN 0883-7252
Preview |
PDF (Accepted manuscript)
- Accepted Version
Download (379kB) | Preview |
Abstract
This paper provides consistent information criteria for the selection of forecasting models which use a subset of both the idiosyncratic and common factor components of a big dataset. This hybrid model approach has been explored by recent empirical studies to relax the strictness of pure factor-augmented model approximations, but no formal model selection procedures have been developed. The main difference to previous factor-augmented model selection procedures is that we must account for estimation error in the idiosyncratic component as well as the factors. Our main contribution is to show the conditions required for selection consistency of a class of information criteria which reflect this additional source of estimation error. We show that existing factor-augmented model selection criteria are inconsistent in circumstances where N is of larger order than root-T, where N and T are the cross-section and time series dimensions of the dataset respectively, and that the standard BIC is inconsistent regardless of the relationship between N and T. We therefore propose a new set of information criteria which guarantee selection consistency in the presence of estimated idiosyncratic components. The properties of these new criteria are explored through a Monte Carlo simulation study. The paper concludes with an empirical application to long-horizon exchange rate forecasting using a recently proposed model with country-specific idiosyncratic components from a panel of global exchange rates.
Item Type: | Article |
---|---|
Uncontrolled Keywords: | forecasting,factor model,model selection,information criteria,idiosyncratic |
Faculty \ School: | Faculty of Social Sciences > School of Economics |
Depositing User: | Pure Connector |
Date Deposited: | 08 Feb 2017 04:21 |
Last Modified: | 22 Oct 2022 02:13 |
URI: | https://ueaeprints.uea.ac.uk/id/eprint/62351 |
DOI: | 10.1002/jae.2567 |
Downloads
Downloads per month over past year
Actions (login required)
![]() |
View Item |