Robust Criteria

摘要

The problem of testing the hypothesis H_0:θ = θ_0 against the alternative hypothesis H_1:θ_1=θ_0+Δ(N~(1/2))~(-1) is examined for the Shurygin noise model, where θ = T(F_θ) is an unknown parameter. In the present report, a class of criteria robust in the sense of the significance value and power density function is considered for parametric andnonparametric problem formulation on statistics of the form S_N = N~(1/2) (V~(1/2)(θ_N))/(θ_n-θ), where V(θ_N) is the θ_N variance andθ_N = T(F_N) is the robust parametric or nonparametric estimate of the parameter θ = T(F_θ ) obtained by the weighted maximum likelihood method.