Factor‐augmented Error Correction Models*

Abstract

This paper brings together several important strands of the econometrics literature: errorcorrection, cointegration and dynamic factor models. It introduces the Factor-augmented Error Correction Model (FECM), where the factors estimated from a large set of variables in levels are jointly modelled with a few key economic variables of interest. With respect to the standard ECM, the FECM protects, at least in part, from omitted variable bias and the dependence of cointegration analysis on the specific limited set of variables under analysis. It may also be in some cases a refinement of the standard Dynamic Factor Model since it allows us to include the error correction terms into the equations, and by allowing for cointegration prevent the errors from being non-invertible moving average processes. In addition, the FECM is a natural generalization of factor augmented VARs (FAVAR) considered by Bernanke, Boivin and Eliasz (2005) inter alia, which are specified in first differences and are therefore misspecified in the presence of cointegration.The FECM has a vast range of applicability. A set of Monte Carlo experiments and two detailed empirical examples highlight its merits in finite samples relative to standard ECM and FAVAR models. The analysis is conducted primarily within an in-sample framework, although the out-ofsample implications are also explored.

Global commodity cycles and linkages: a FAVAR approach

“…Given that the variables included in the VAR are all integrated, as indicated by unit root tests, we also test for cointegration, to see if using a FAVECM (Banerjee and Marcellino 2008) or a FAVAR in first differences (Bernanke et al 2005) is the most appropriate procedure. We extract two common factors and see that their loadings are clustered in such a way that the first factor appears to be a "metals" factor and the second one, a predominantly "food" factor.…”

“…We extract two common factors and see that their loadings are clustered in such a way that the first factor appears to be a "metals" factor and the second one, a predominantly "food" factor. In the second step, we estimate a FAVECM model (Banerjee and Marcellino 2008) for each non-energy commodity price, including the fundamentals and the common factors and test for cointegration. As we find no evidence of cointegration in this model, we conduct the analysis with a FAVAR specified in growth rates.…”

Mining Big Data Using Parsimonious Factor and Shrinkage Methods

“…Armah and Swanson (2010a,b), Artis et al (2005), Bai and Ng (2002, 2006b, 2008, Ng (2005, 2006), Ding and Hwang (1999), Stock and Watson (2002a). Other recent important papers which extend the discussion in the above papers to vector and error-correction type models include Banerjee and Marcellino (2008), Dufour and Stevanovic (2010).…”

Short-Term Forecasting for Empirical Economists. A Survey of the Recently Proposed Algorithms

“…14 Poncela and Ruiz (2012) show that, adding an additional predictor, the …ltered and the smoothed steady-state MSE typically decrease. They remain constant only if (i) the common factor is static, 11 = 0; or (ii) the additional predictor is not informative, N +1 = 0 and its corresponding idiosyncratic noise, u N +1 ; is not correlated with any of the predictors already included in the system.…”