Stable tail dependence functions -- some basic properties
The paper discusses multivariate extreme value (MEV) distributions and their associated stable tail dependence functions (STDFs), which are used to characterize the dependence structure of extreme events in multivariate data. The STDFs are defined as a function of the distribution function and have important properties such as being homogeneous, normalized, and fully \(d\)-alternating. The article focuses on two main topics: the extremal coefficients of a \(d\)-variate STDF and logistic, negative logistic, and nested logistic STDFs. Theorems are presented to provide new insights into the properties of these functions and their relationship to the independence of sub-vectors. The paper also provides recommendations for further reading on STDFs. In Section 2, the author starts by introducing a notation for multivariate real-valued functions, next, the author defines the concept of fully \(d\)-alternating functions to be functions that satisfy certain monotonicity conditions and then he defines the set of fully \(d\)-alternating, homogeneous, and normalized functions on the positive orthant (a mathematical object consisting of points with non-negative coordinates). In Section 3, the author discusses some properties of STDFs (standardized tail dependence functions) and their relation to MEVs (max-stable random vectors) and their distributions. Section 4 focuses on characterizing logistics and related STDF's. The author further discusses stable tail dependence functions (STDFs), which are functions that describe the probability of observing extreme values of two random variables simultaneously. The logistic STDFs are discussed in detail. The author concludes that among all symmetric STDFs, the logistic STDFs are particular, and there are no other STDFs with this property besides the logistic ones. Finally, a proof is provided by the author for a theorem that characterizes logistic STDFs.
- Copulas, stable tail dependence functions, and multivariate monotonicity
- A directory of coefficients of tail dependence
- Tail dependence from a distributional point of view
- Almost sure convergence of the stable tail empirical dependence function in multivariate extreme statistics
- Best attainable rates of convergence for estimators of the stable tail dependence function
- A characterization of Gumbel's family of extreme value distributions
- A multivariate version of Williamson's theorem, \(\ell^1\)-symmetric survival functions, and generalized Archimedean copulas
- An introduction to copulas.
- Characterizations of a multivariate extreme value distribution
- Copulas, stable tail dependence functions, and multivariate monotonicity
- Functions operating on multivariate distribution and survival functions - With applications to classical mean-values and to copulas
- Homogeneous distributions -- and a spectral representation of classical mean values and stable tail dependence functions
- Some properties of multivariate extreme value distributions and multivariate tail equivalence
- Statistics of Extremes
This page was built for publication: Stable tail dependence functions -- some basic properties
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2172583)
