Factor models are becoming increasingly popular in economics because they can utilize large data sets in an effective manner. Factor models have been used for various purposes. First, they have been used to construct economic indicators. Monthly coincident business cycle indicators such as the Chicago Fed National Activity index (CFNAI) for the US and EuroCOIN for the Euro area (cf. Altissimo et al., 2001) are related examples. Second, factor models have widely been used in order to forecast real and nominal economic variables. They often provide more accurate forecasts than autoregressive and vector autoregressive models (see Eickmeier and Ziegler, 2008 and the literature cited therein). Third, factor models have been used for monetary policy analysis in combination with a vector autoregressive (VAR) system as in Bernanke et al. (2005). In many cases only five to ten factors are sufficient to capture more than a half of the total variation within a data set of more than three hundred macroeconomic variables. Therefore, adding a few common factors to a macroeconomic VAR system is supposed to control for a variety of omitted variables within a typical low-dimensional VAR analysis. Fourth, factor models are used for instrumental variables estimation. Bai and Ng (2010) assume that endogenous regressors are driven by a small number of unobserved, exogenous factors and suggest using the estimated factors as instruments. Fifth, factor models have been used in panel regressions as a way of modelling cross-sectional correlation.
You are not authenticated to view the full text of this chapter or article.
Get access to the full article by using one of the access options below.