In this paper we shall give an alternative derivation of the coe cient of tail dependence introduced by Ledford and Tawn [1996, Biometrika 83, 169 -187] and propose a consistent estimator, which is asymptotically normal.
β¦ LIBER β¦
On the impossibility of estimating densities in the extreme tail
β Scribed by Jan Beirlant; Luc Devroye
- Publisher
- Elsevier Science
- Year
- 1999
- Tongue
- English
- Weight
- 100 KB
- Volume
- 43
- Category
- Article
- ISSN
- 0167-7152
No coin nor oath required. For personal study only.
β¦ Synopsis
We give a short proof of the following result. Let X1; : : : ; Xn be independent and identically distributed observations drawn from a density f on the real line. Let fn be any estimate of the density gn of max(X1; : : : ; Xn). We show that there exists a unimodal inΓΏnitely many times di erentiable density f such that
Thus, in the total variation sense, universally consistent density estimates do not exist. A similar result is derived concerning the supremum norm.
π SIMILAR VOLUMES
Estimation of the coefficient of tail de
β
L. Peng
π
Article
π
1999
π
Elsevier Science
π
English
β 137 KB
Estimation of the tail parameter in the
β
M.I. Fraga Alves
π
Article
π
1995
π
Elsevier Science
π
English
β 880 KB
On the estimation of the Weibull tail co
β
Michel Broniatowski
π
Article
π
1993
π
Elsevier Science
π
English
β 789 KB
The efficient estimation of tail probabi
β
Paul Blackwell
π
Article
π
1994
π
Springer US
π
English
β 454 KB
A new approach on estimation of the tail
β
Dimitris N. Politis
π
Article
π
2002
π
Elsevier Science
π
English
β 56 KB
On the estimation of the quantile densit
β
Michael Falk
π
Article
π
1986
π
Elsevier Science
π
English
β 260 KB