Dependence structures for multivariate high-frequency data in finance

By Wolfgang Breymann, Alexandra Dias, Paul Embrechts. 2003

527 citations recorded by the source corpus.

Stylized facts for univariate high-frequency data in finance are well known. They include scaling behaviour, volatility clustering, heavy tails and seasonalities. The multivariate problem, however, has scarcely been addressed up to now. In this paper, bivariate series of high-frequency FX spot data for major FX markets are investigated. First, as an indispensable prerequisite for further analysis, the problem of simultaneous deseasonalization of high-frequency data is addressed. In the following sections we analyse in detail the dependence structure as a function of the timescale. Particular emphasis is put on the tail behaviour, which is investigated by means of copulas.

Subfields: Artificial Intelligence, Economics and Econometrics, Finance

View Paper: https://www.semanticscholar.org/paper/86b335ac2a0dac0b2d40a3eabbfaeb5597a57378

How researchers see this paper

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

In Artificial Intelligence

In Economics and Econometrics

In Finance

Citations in the corpus

Copula goodness-of-fit testing: an overview and power comparison (2009, methodology): “Lately, several contributions have been made to test this hypothesis, e.g. Genest and Rivest (1993), Shih (1998), Breymann et al. (2003), Malevergne and Sornette (2003), Scaillet (2005), Genest and Rémillard (2008), Fermanian (2005), Panchenko (2005), Berg and Bakken (2005), Genest et al. (2006a),…”
Testing Goodness of Fit for Parametric Families of Copulas—Application to Financial Data (2005, background): “Various authors have addressed this question (see e.g., Breymann et al., 2003; Fermanian, 2003; Junker and May, 2002; Malevergne and Sornette, 2003; Mashal and Zeevi, 2002; Savu and Trede, 2004).”
Copula-based multivariate GARCH model with uncorrelated dependent errors (2009, methodology): “For example, taking the empirical distribution functions for the margins and a parametric function for the copula, Breymann et al. (2003) and Chen and Fan (2006) estimate D2t ≡ diag (Ht) using univariate realized volatility estimated from high frequency data or using univariate GARCH models.”
A framework of change‐point detection for multivariate hydrological series (2015, methodology): “For the trend case, Chebana et al. [2013] employed the nonparametric Mann-Kendall type and Spearman’s rho type tests to detect the trends in multivariate hydrological series.”
Self‐consistent method for density estimation (2009, methodology): “Among many possible implementations, the method could then be applied to the analysis of multielectrode neuronal recordings (Brown et al., 2004), multivariate financial data (Breymann et al., 2003) and reconstruction of Ramachandran angle distributions (Kleywegt and Jones, 1996).”
The t Copula and Related Copulas (2005, methodology): “This finding underscores an empirical observation by Breymann et al. (2003) that for bivariate financial return data where the t copula seemed to be the best overall copula model for the dependence, the Clayton copula seemed to be the best model for the most extreme observations in the joint lower…”
Risk management with high-dimensional vine copulas: An analysis of the Euro Stoxx 50 (2013, methodology): “, Mashal and Zeevi (2002) and Breymann et al. (2003)). The Gaussian copula cannot account for tail dependence, but it is nevertheless still frequently used and therefore fitted for comparison. The parameters of the two elliptical copulas are also estimated by maximum likelihood estimation. Results a”
Multivariate non-normally distributed random variables in climate research - introduction to the copula approach (2008, methodology): “In the approach proposed by (Breymann et al., 2003) the PIT is performed by deﬁning a univariate random vector Under H 0 it follows a χ 2 -distribution with m degrees of freedom, so that W ∗ = F χ 2 m (Y ∗ ) should be a univariate random variable from U ( 0 , 1 ) .”