ISSN:
1572-915X
Keywords:
extreme value distribution
;
generalized Pareto distribution
;
local asymptotic normality
;
asymptotically uniformly optimal test
;
asymptotic envelope power function
Source:
Springer Online Journal Archives 1860-2000
Topics:
Mathematics
Notes:
Abstract We consider the full statistical families of extreme value distributions $$G_{\beta ,\sigma ,\mu } $$ and generalized Pareto distributions $$H_{\beta ,\sigma ,\mu } $$ , where $$\beta \in \mathbb{R},\sigma 〉 0$$ , and $$\mu \in \mathbb{R}$$ denote the shape, scale and location parameters, respectively. We consider the testing problems $$\beta \geqslant 0$$ against $$\beta 〈 0$$ and β = 0 against β ≠ 0, where σ and μ are treated as nuisance parameters. Showing local asymptotic normality (LAN), we derive asymptotic envelope power functions for test sequences and establish tests which attain these upper bounds. The finite sample size behavior is studied by simulations.
Type of Medium:
Electronic Resource
URL:
http://dx.doi.org/10.1023/A:1012209826248
Permalink