Forecasting Conditional Covariance Matrices in High-Dimensional Time Series: A General Dynamic Factor Approach

Trucíos, Carlos; Mazzeu, João Henrique Gonçalves; Hallin, Marc; Hotta, Luiz Koodi; Pereira, Pedro L. Valls; Zevallos, Mauricio

doi:10.2139/ssrn.3399782

Cited by 2 publications

(3 citation statements)

References 71 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For these reasons, Barigozzi and Hallin (2016) introduce a two-step GDFM approach by which the nonparametric and model-free virtues of factor models are used in a joint analysis of returns and volatilities. In Barigozzi and Hallin (2017a), that two-step GDFM is combined with a GARCH strategy in order to produce point-forecasts for volatilities (see also Trucíos et al, 2019 for a recent example), while Barigozzi and Hallin (2017b) and Barigozzi et al (2018) apply the same methodology in a study of the dynamic interdependencies of US and international financial markets. A two-stage factor approach similar to ours but in a static factor model setting is proposed in Chicheportiche and Bouchaud (2015).…”

Section: Introductionmentioning

confidence: 99%

Generalized dynamic factor models and volatilities: Consistency, rates, and prediction intervals

Barigozzi

Hallin

2020

Journal of Econometrics

Self Cite

View full text Add to dashboard Cite

Volatilities, in high-dimensional panels of economic time series with a dynamic factor structure on the levels or returns, typically also admit a dynamic factor decomposition. We consider a two-stage dynamic factor model method recovering the common and idiosyncratic components of both levels and log-volatilities. Specifically, in a first estimation step, we extract the common and idiosyncratic shocks for the levels, from which a log-volatility proxy is computed. In a second step, we estimate a dynamic factor model, which is equivalent to a multiplicative factor structure for volatilities, for the log-volatility panel. By exploiting this two-stage factor approach, we build one-step-ahead conditional prediction intervals for large n × T panels of returns. Those intervals are based on empirical quantiles, not on conditional variances; they can be either equal-or unequaltailed. We provide uniform consistency and consistency rates results for the proposed estimators as both n and T tend to infinity. We study the finite-sample properties of our estimators by means of Monte Carlo simulations. Finally, we apply our methodology to a panel of asset returns belonging to the S&P100 index in order to compute one-step-ahead conditional prediction intervals for the period 2006-2013. A comparison with the componentwise GARCH benchmark (which does not take advantage of cross-sectional information) demonstrates the superiority of our approach, which is genuinely multivariate (and high-dimensional), nonparametric, and model-free.JEL Classification: C32, C38, C58.

show abstract

Section: Introductionmentioning

confidence: 99%

Generalized dynamic factor models and volatilities: Consistency, rates, and prediction intervals

Barigozzi

Hallin

2020

Journal of Econometrics

Self Cite

View full text Add to dashboard Cite

show abstract

“…The factors, being estimated from the high dimensional data, can be used for either descriptive or predictive purposes. Applications include: forecasting macroeconomic time series (Stock and Watson, 2002a,b;Forni et al, 2005;Bai and Ng, 2008); excess returns in stock and bond markets Ng, 2007, 2009); construction of business cycle indicators and nowcasting (Cristadoro et al, 2005;Giannone et al, 2008;Altissimo et al, 2010); structural macroeconomic analysis and monetary policy (Bernanke and Boivin, 2003;Favero et al, 2005;Stock and Watson, 2005;Eickmeier, 2007;Forni et al, 2009;Forni and Gambetti, 2010); prediction of conditional variance-covariance matrix (Alessi et al, 2009;Aramonte et al, 2013;Trucíos et al, 2019b), to quote only a few.…”

Section: Introductionmentioning

confidence: 99%

“…The Forni et al (2015Forni et al ( , 2017 procedure has been successfully used to forecast inflation and financial returns; see Della Marra (2017), Forni et al (2018), Giovannelli et al (2018). It also has been used in the prediction of the conditional variances of financial returns, the extraction of market shocks (Barigozzi and Hallin, 2016, and the prediction of conditional variance-covariance matrices (Trucíos et al, 2019b).…”

Section: Introductionmentioning

confidence: 99%

Robustness and the General Dynamic Factor Model With Infinite-Dimensional Space: Identification, Estimation, and Forecasting

et al. 2019

Self Cite

View full text Add to dashboard Cite

Os artigos dos Textos para Discussão da Escola de Economia de São Paulo da Fundação GetulioVargas são de inteira responsabilidade dos autores e não refletem necessariamente a opinião da FGV-EESP. É permitida a reprodução total ou parcial dos artigos, desde que creditada a fonte. AbstractGeneral dynamic factor models have demonstrated their capacity to circumvent the curse of dimensionality in time series and have been successfully applied in many economic and financial applications. However, their performance in the presence of outliers has not been analysed yet. In this paper, we 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 identification, estimation and forecasting procedures. Our proposal is evaluated via Monte Carlo experiments and in empirical data.

show abstract

Forecasting Conditional Covariance Matrices in High-Dimensional Time Series: A General Dynamic Factor Approach

Cited by 2 publications

References 71 publications

Generalized dynamic factor models and volatilities: Consistency, rates, and prediction intervals

Generalized dynamic factor models and volatilities: Consistency, rates, and prediction intervals

Robustness and the General Dynamic Factor Model With Infinite-Dimensional Space: Identification, Estimation, and Forecasting

Contact Info

Product

Resources

About