### Abstract

Original language | English |
---|---|

Place of Publication | Tilburg |

Publisher | CentER, Center for Economic Research |

Number of pages | 32 |

Volume | 2020-004 |

Publication status | Published - 3 Feb 2020 |

### Publication series

Name | CentER Discussion Paper |
---|---|

Volume | 2020-004 |

### Fingerprint

### Keywords

- extreme value statistics
- functional data
- tail empirical process
- tal dependence
- tial copula estimation
- uniform asymptotic normality

### Cite this

*Empirical Tail Copulas for Functional Data*. (CentER Discussion Paper; Vol. 2020-004). Tilburg: CentER, Center for Economic Research.

}

**Empirical Tail Copulas for Functional Data.** / Einmahl, John; Segers, Johan.

Research output: Working paper › Discussion paper › Other research output

TY - UNPB

T1 - Empirical Tail Copulas for Functional Data

AU - Einmahl, John

AU - Segers, Johan

N1 - CentER Discussion Paper Nr. 2020-004

PY - 2020/2/3

Y1 - 2020/2/3

N2 - For multivariate distributions in the domain of attraction of a max-stable distribution, the tail copula and the stable tail dependence function are equivalent ways to capture the dependence in the upper tail. The empirical versions of these functions are rank-based estimators whose inflated estimation errors are known to converge weakly to a Gaussian process that is similar in structure to the weak limit of the empirical copula process. We extend this multivariate result to continuous functional data by establishing the asymptotic normality of the estimators of the tail copula, uniformly over all finite subsets of at most D points (D fixed). As a special case we obtain the uniform asymptotic normality of all estimated upper tail dependence coefficients. The main tool for deriving the result is the uniform asymptotic normality of all the D-variate tail empirical processes. The proof of the main result is non-standard.

AB - For multivariate distributions in the domain of attraction of a max-stable distribution, the tail copula and the stable tail dependence function are equivalent ways to capture the dependence in the upper tail. The empirical versions of these functions are rank-based estimators whose inflated estimation errors are known to converge weakly to a Gaussian process that is similar in structure to the weak limit of the empirical copula process. We extend this multivariate result to continuous functional data by establishing the asymptotic normality of the estimators of the tail copula, uniformly over all finite subsets of at most D points (D fixed). As a special case we obtain the uniform asymptotic normality of all estimated upper tail dependence coefficients. The main tool for deriving the result is the uniform asymptotic normality of all the D-variate tail empirical processes. The proof of the main result is non-standard.

KW - extreme value statistics

KW - functional data

KW - tail empirical process

KW - tal dependence

KW - tial copula estimation

KW - uniform asymptotic normality

M3 - Discussion paper

VL - 2020-004

T3 - CentER Discussion Paper

BT - Empirical Tail Copulas for Functional Data

PB - CentER, Center for Economic Research

CY - Tilburg

ER -