Modelling and forecasting multivariate realized volatility

Abstract

Standard-Nutzungsbedingungen:Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden.Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen.Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in EconStor may University of Konstanz CoFEValeri Voev ‡ University of Aarhus CREATESAbstract This paper proposes a methodology for modelling time series of realized covariance matrices in order to forecast multivariate risks. The approach allows for flexible dynamic dependence patterns and guarantees positive definiteness of the resulting forecasts without imposing parameter restrictions. We provide an empirical application of the model, in which we show by means of stochastic dominance tests that the returns from an optimal portfolio based on the model's forecasts second-order dominate returns of portfolios optimized on the basis of traditional MGARCH models. This result implies that any risk-averse investor, regardless of the type of utility function, would be better-off using our model.JEL classification: C32, C53, G11

The Economic Value of Volatility Forecasts: A Conditional Approach

“…Given the documented benefits of vector autoregressive (VAR) models of realised volatility (see, e.g., Andersen et al, 2003), the next method we consider assumes that forecasts are given 3 The covariance matrix measure used is based on the subsampling methodology of Zhang et al (2005) within a multivariate setting (see, e.g., Chiriac and Voev, 2011, for use of this methodology in a similar context). Further details are provided in Appendix A.…”

“…Commonly-employed examples include the Sharpe ratio and the performance fee associated with conditionally-optimised portfolios constructed using volatility forecasts; see, e.g., West et al (1993), Fleming et al (2001Fleming et al ( , 2003, Marquering and Verbeek (2004), Şerban et al (2007), Clements et al (2009), Liu (2009, and Chiriac and Voev (2011). 1 Such measures are unconditional in nature and deliver a single performance measure over the entire sample.…”

The Realized Hierarchical Archimedean Copula in Risk Modelling

“…The formal definition of the realized measures is given in Appendix B. Despite the constantly growing research on incorporating the realized measures into multivariate Gaussian models, discussed in Chiriac and Voev (2011) and Bauer and Vorkink (2011), and into GARCH type models, for example, Hansen et al (2014) and Bollerslev et al (2016), there is still a gap in the literature on how the parameters of non-Gaussian copula can be estimated daily based on high-frequency observations. It is important to note here that such standard copula estimation techniques as the Maximum Likelihood (ML) method or the inversion of Kendall's τ can not be directly applied to tick-by-tick observations.…”

“…Many researchers have implemented the obtained realized measures to model financial time series. Most of those studies, however, employ models where the realized correlation matrix directly characterizes the multivariate distribution, see, for example, Bauer and Vorkink (2011), Chiriac and Voev (2011), Jin and Maheu (2012), or address GARCH type models, for example, Hansen et al (2014), Bauwens et al (2012), Noureldin et al (2012), Bollerslev et al (2016). There are only a limited number of studies which discuss the implementation of high-frequency data in copula models.…”

Estimation and Forecasting of Large Realized Covariance Matrices and Portfolio Choice

“…Furthermore, the parameters of their model are not easily interpretable because of the log-matrix transformation, and thus one can not investigate the main driving forces of the dynamics of the covariance matrix as in our approach. Another example is Chiriac and Voev (2011) who consider a multivariate ARFIMA model to forecast realized measures of daily covariance matrices. As in Bauer and Vorkink (2011), only small covariance matrices can be modelled.…”