Probabilistic models

On numerical approximation of tests' power for the case of Laplace distribution

R. A. Korolev

Lomonosov Moscow State University, Faculty of Computational Mathematics and Cybernetics

Abstract: In the paper we obtain numerical results for the power of the asymptotically optimal test and for the power of the asymptotically most powerful test for the case of Laplace distribution (see [1], [2]). For both tests we investigate the numerical accuracy of approximations to power functions using their asymptotic expansions of the order $n^{-1}$ (see [2], Theorem 2.1, Theorem 3.3). We numerically research deficiency of the asymptotically optimal test.

Keywords: asymptotic expansion, numerical approximation, power function, Laplace or double exponential distribution, deficiency.

UDC: 519.2

Received: 29.01.2009
Revised: 17.02.2009

Bibliographic databases:

•