Detecting big structural breaks in large factor models

By Liang Chen, Juan J. Dolado, J. Gonzalo. 2014

136 citations recorded by the source corpus.

Time invariance of factor loadings is a standard assumption in the analysis of large factor models. Yet, this assumption may be restrictive unless parameter shifts are mild (i.e., local to zero). In this paper we develop a new testing procedure to detect big breaks in these loadings at either known or unknown dates. It relies upon testing for parameter breaks in a regression of one of the factors estimated by Principal Components analysis on the remaining estimated factors, where the number of factors is chosen according to Bai and Ng’s (2002) information criteria. The test fares well in terms of power relative to other recently proposed tests on this issue, and can be easily implemented to avoid forecasting failures in standard factor-augmented (FAR, FAVAR) models where the number of factors is a priori imposed on the basis of theoretical considerations.

Subfields: Economics and Econometrics, Finance

View Paper: https://www.semanticscholar.org/paper/2e006de32d6028e95024a8af64b56ea0fb792ad1

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Economics and Econometrics

In Finance

Citations in the corpus

On Time-Varying Factor Models: Estimation and Testing ∗ (2017, methodology): “The tests of Chen et al. (2014) and Han and Inoue (2014) involve the long run variance estimation.”
Testing for Factor Loading Structural Change under Common Breaks (2015, background): “3Recently, two other factor loading structural change tests have been proposed by Chen et al. (2012) and Han and Inoue (2013).”
Testing for Structural Stability of Factor Augmented Forecasting Models (2014, background): “Clements and Hendry 2002, Hendry and Mizon 2005, and Castle, Doornik and Hendry 2010). Among the main causes of instability, Hendry and Clements (2002) point out the importance of intercept shifts, mainly arising because of shifts in the means of omitted variables.”
The Effects of Global Shocks on Small Commodity-Exporting Economies: Lessons from Canada (2014, unknown): “To do so, we have performed the time invariance test of the factor loading proposed by Chen, Dolado, and Gonzalo (2014) for the Canadian block of the SDFM model and found some indication of a structural break in the early 1990s.29 Additionally, we analyzed possible breakpoints in the individual VAR…”
Dynamic Factor Models, Factor-Augmented Vector Autoregressions, and Structural Vector Autoregressions in Macroeconomics (2016, unknown): “Tests proposed by Chen et al. (2014) and Han and Inoue (2015) test for a general break in Λ (all equations) by noting that, if Λ changes, the covariance matrix of the full-sample PC estimator will change at the break date in Λ.”
Simultaneous multiple change-point and factor analysis for high-dimensional time series (2016, unknown): “While the majority of change-point methods for factor models rely on the consistent estimation of the number of factors (Breitung and Eickmeier, 2011; Han and Inoue, 2014; Chen et al., 2014; Corradi and Swanson, 2014), our empirical study on simulated data, available in the supplementary document, i”
Identification and estimation of a large factor model with structural instability (2017, methodology): “Compared to √ T N → 0, which is assumed in Chen et al. (2014), Han and Inoue (2015), Assumption 9 is significantly weaker and much easier to be satisfied since even when T is much larger than N , log T N could still be very close to zero.”
TESTS FOR PARAMETER INSTABILITY IN DYNAMIC FACTOR MODELS (2013, methodology): “…shocks when the Heteroskedasticity and Autocorrelation Consistent (henceforth HAC) covariance matrix estimator is used.2 More recently, Chen, Dolado, and Gonzalo (2012) (CDG hereafter) develop Wald and LM tests of structural change in factor loadings using a regression of the first…”
Estimation of large dimensional factor models with an unknown number of breaks (2018, methodology): “As shown in a recent paper of Chen et al. (2014), their Sup-Wald test detected one break date around 1979-1980 (second oil price shock).”