KENDALL'S TAU FOR ELLIPTICAL DISTRIBUTIONS ∗

By Filip Lindskog, Alexander J. McNeil, Uwe Schmock. 2003

212 citations recorded by the source corpus.

Subfields: Economics and Econometrics, Finance, Global and Planetary Change

View Paper: https://www.semanticscholar.org/paper/a73406353f97ed15589daeeb7c2020e668581145

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

In Global and Planetary Change

Citations in the corpus

Extensions to the Gaussian Copula: Random Recovery and Random Factor Loadings (2005, background): “The equality τ(Yi,Xj) = 2π−1 sin−1(bi ·a j/σi) is a consequence of a general result for elliptical distributions in Lindskog et al. (2003); the equation τ(Ri,Xj) = τ(Yi,Xj) follows from the fact that Kendall’s Tau is invariant under strictly increasing transformations (see e.g. Embrechts et al.…”
Analysis of elliptical copula correlation factor model with Kendall's tau (2013, methodology): “, we can relate the elements of the copula correlation matrix Σ to the elements of the Kendall’s tau matrix T independently of the characteristic generator via the formula Σ=sin π 2 T ; (1.4) see [14, 21, 22, 26, 27]. Here and throughout the paper, we use the convention that the sign, sine and cos”
Dependence structures for multivariate high-frequency data in finance (2003, methodology): “is more efficient and robust than Pearson’s linear correlation estimator; see Lindskog et al. (2001). In order to estimate the ratio between the standard deviations we can use another dispersion estimator rather than the standard one, S2 n = ∑n i=1(xi − x)2/n, like the median absolute deviation, mad”
Estimating the tail-dependence coefficient: Properties and pitfalls (2005, methodology): “Then the EV distribution possesses the following representation, Pickands (1981): FEV (x, y) = exp { − ( 1 x + 1 y ) A ( y x + y )} , x, y > 0.”
Dependence Properties of Meta-Elliptical Distributions (2005, result): “is independent of g. This result is reported also by Lindskog, McNeil and Schmock (2001) and Frahm, Junker and Szimayer (2003).”
Bayesian inference for multivariate copulas using pair-copula constructions. (2010, background): “A key role plays here the following relationship τ = 2 π · arcsin ρ (4.2) proven by Lindskog et al. (2003) for elliptical distributions with continuous margins.”
Risk management with high-dimensional vine copulas: An analysis of the Euro Stoxx 50 (2013, methodology): “The parameters of the Student-t copula are estimated in two steps (Lindskog et al. 2003).”
The t Copula and Related Copulas (2005, methodology): “This fact is much more widely known and has been demonstrated in a variety of different ways (see Sibuya (1961) or Resnick (1987), Chapter 5).”
Empirical estimation of tail dependence using copulas: application to Asian markets (2005, unknown): “For this symmetrical copula we have a relationship between and Kendall’s tau, see Lindskog et al. (2001): ðZ1,Z2Þ ¼ 2 arcsinð ðZ1,Z2ÞÞ: ð2Þ Then, the lower and upper tail dependence coefficients for the Student copula are given by the following equation: L ¼ U ¼ 2t…”
Modeling exchange rate dependence dynamics at different time horizons (2010, methodology): “Since for our data the best performing copula models belong to the elliptical family we transform Kendall's tau estimates, τ Kendall , into linear correlation, ρ τ , using the relationship τ Kendall,t = 2 π arcsin ρ τ,t , (11) which is valid for elliptical distributions; see Lindskog et al. (2003).”