scientific article; zbMATH DE number 3860170
[For the entire collection see Zbl 0504.00022.] Let \(X_ 1,...,X_ n\) be independent random variables, identically distributed according to a distribution function F(x-\(\theta)\) with an unknown parameter \(\theta\), where F is a continuous distribution function satisfying \(F(x)+F(-x)=1\) and \(0<F(x)<1 (x\in R^ 1)\). The author considers the problem of testing the hypothesis \(H:\theta =\theta_ 0\) against the alternative \(K:\theta>\theta_ 0.\) The power function of a consistent test tends to 1 as \(\theta -\theta_ 0\to \infty\). He considers the rate of this convergence and shows that the rate at which the tails of the distribution F tend to 0 form a natural upper bound on the rate of convergence of the power function of any monotone test. More precisely, he shows that the probability of the error of the second kind tends to 0 at most n-times faster than the tails of the distribution.
- scientific article; zbMATH DE number 3844816
- On a power of optimal test for asymptotic distinction of statistical hypotheses for distributions with heavy tails
- scientific article; zbMATH DE number 3960756
- Locally most powerful rank tests
- Asymptotic distribution of the most powerful invariant test for invariant families
This page was built for publication:
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3327503)