作者: Karl-Heinz Jockel
关键词:
摘要: Since their introduction by Dwass (1957) and Barnard (1963), Monte Carlo tests have attracted considerable attention. The aim of this paper is to give a unified approach that covers the case an arbitrary null distribution in order study statistical properties under hypothesis alternative. For finite samples we obtain bounds for power test with original allow determination required simulation effort. Furthermore concept asymptotic (resp. local asymptotic) relative Pitman efficiency (ARPE, resp. LARPE) adapted behaviour. normal limit investigated more detail, leading explicit formulas ARPE LARPE.