Robustness and the general dynamic factor model with infinite-dimensional space: identification, estimation, and forecasting

Abstract General dynamic factor models have demonstrated their capacity to circumvent the curse of dimensionality in the analysis of high-dimensional time series and have been successfully considered in many economic and financial applications. As second-order models, however, they are sensitive to the presence of outliers—an issue that has not been analyzed so far in the general case of dynamic factors with possibly infinite-dimensional factor spaces (Forni et al. 2000, 2015, 2017). In this paper, we consider this robustness issue and study the impact of additive outliers on the identification, estimation, and forecasting performance of general dynamic factor models. Based on our findings, we propose robust versions of identification, estimation, and forecasting procedures. The finite-sample performance of our methods is evaluated via Monte Carlo experiments and successfully applied to a classical data set of 115 US macroeconomic and financial time series.

[1]  J. Bai,et al.  Determining the Number of Factors in Approximate Factor Models , 2000 .

[2]  M. Hallin,et al.  Dynamic factor models with infinite-dimensional factor spaces: One-sided representations , 2013 .

[3]  M. Genton,et al.  Highly Robust Estimation of the Autocovariance Function , 2000 .

[4]  Marco Lippi,et al.  The general dynamic factor model: One-sided representation results , 2011 .

[5]  Mia Hubert,et al.  ROBPCA: A New Approach to Robust Principal Component Analysis , 2005, Technometrics.

[6]  Carlos Trucíos,et al.  Forecasting Conditional Covariance Matrices in High-Dimensional Time Series: A General Dynamic Factor Approach , 2019, SSRN Electronic Journal.

[7]  J. Stock,et al.  Macroeconomic Forecasting Using Diffusion Indexes , 2002 .

[8]  David H. Small,et al.  Nowcasting: the real time informational content of macroeconomic data releases , 2008 .

[9]  P. Rousseeuw,et al.  The Change-of-Variance Curve and Optimal Redescending M-Estimators , 1981 .

[10]  Peng Wang,et al.  Econometric Analysis of Large Factor Models , 2016 .

[11]  P. Rousseeuw,et al.  Alternatives to the Median Absolute Deviation , 1993 .

[12]  J. Stock,et al.  Forecasting Using Principal Components From a Large Number of Predictors , 2002 .

[13]  Peter J. Rousseeuw,et al.  Fast Robust Correlation for High-Dimensional Data , 2017, Technometrics.

[14]  M. Hallin,et al.  The Generalized Dynamic-Factor Model: Identification and Estimation , 2000, Review of Economics and Statistics.

[15]  Fabio Della Marra.  A forecasting performance comparison of dynamic factor models based on static and dynamic methods , 2017 .

[16]  M. Hubert,et al.  A fast method for robust principal components with applications to chemometrics , 2002 .

[17]  Ricardo A. Maronna,et al.  Principal Components and Orthogonal Regression Based on Robust Scales , 2005, Technometrics.

[18]  Jason Wu,et al.  Dynamic Factor Value-at-Risk for Large Heteroskedastic Portfolios , 2012 .

[19]  R. Baragona,et al.  Outliers in dynamic factor models , 2007, 0710.3676.

[20]  M. Hallin,et al.  Identification of Global and Local Shocks in International Financial Markets via General Dynamic Factor Models , 2019 .

[21]  Mia Hubert,et al.  MacroPCA: An All-in-One PCA Method Allowing for Missing Values as Well as Cellwise and Rowwise Outliers , 2018, Technometrics.

[22]  Marco Lippi,et al.  OPENING THE BLACK BOX: STRUCTURAL FACTOR MODELS WITH LARGE CROSS SECTIONS , 2009, Econometric Theory.

[23]  Carlos Trucíos,et al.  On the robustness of the principal volatility components , 2019 .

[24]  Mario Forni,et al.  The Dynamic Effects of Monetary Policy: A Structural Factor Model Approach , 2008 .

[25]  C. Croux,et al.  Principal Component Analysis Based on Robust Estimators of the Covariance or Correlation Matrix: Influence Functions and Efficiencies , 2000 .

[26]  M. Hallin,et al.  Generalized dynamic factor models and volatilities: estimation and forecasting , 2017 .

[27]  Filippo Altissimo,et al.  New Eurocoin: Tracking Economic Growth in Real Time , 2006, The Review of Economics and Statistics.

[28]  Clifford Lam,et al.  Factor modeling for high-dimensional time series: inference for the number of factors , 2012, 1206.0613.

[29]  R. Tsay,et al.  Outlier Detection in Multivariate Time Series by Projection Pursuit , 2006 .

[30]  L. Guttman.  Some necessary conditions for common-factor analysis , 1954 .

[32]  M. Hallin,et al.  Generalized dynamic factor models and volatilities: Consistency, rates, and prediction intervals , 2018, Journal of Econometrics.

[33]  Sanne Engelen,et al.  A comparison of three procedures for robust PCA in high dimensions , 2016 .

[34]  Marco Lippi,et al.  Dynamic Factor Model with Infinite Dimensional Factor Space: Forecasting , 2016, Journal of Applied Econometrics.

[35]  M. Hallin,et al.  Dynamic Factor Models with Infinite-Dimensional Factor Space: Asymptotic Analysis , 2015 .

[36]  H. Veiga,et al.  Detecting outliers in multivariate volatility models: a wavelet procedure , 2019 .

[37]  M. Hallin,et al.  Determining the Number of Factors in the General Dynamic Factor Model , 2007 .

[38]  Johannes Tang Kristensen.  Factor-based forecasting in the presence of outliers: Are factors better selected and estimated by the median than by the mean? , 2012 .

[39]  Marco Lippi,et al.  The generalized dynamic factor model: consistency and rates , 2004 .

[40]  P. Rousseeuw,et al.  Explicit scale estimators with high breakdown point , 1992 .

[41]  Sydney C. Ludvigson,et al.  Macro Factors in Bond Risk Premia , 2005 .

[42]  Olivier Darné,et al.  Dynamic Factor Models: A Review of the Literature , 2013 .

[43]  Barbara Rossi,et al.  Forecast comparisons in unstable environments , 2010 .

[44]  F. Diebold,et al.  Comparing Predictive Accuracy , 1994, Business Cycles.

[45]  R. Tsay,et al.  Outliers in multivariate time series , 2000 .

[46]  Marco Lippi,et al.  Factor models in high-dimensional time series—A time-domain approach , 2013 .

[47]  D. Massacci,et al.  Forecasting Stock Returns with Large Dimensional Factor Models , 2017 .

[48]  Marco Lippi,et al.  The Generalized Dynamic Factor Model , 2002 .

[49]  J. Stock,et al.  Why Has U.S. Inflation Become Harder to Forecast , 2007 .

[50]  Andrés M. Alonso,et al.  A robust procedure to build dynamic factor models with cluster structure , 2020 .