| 摘要: | 考慮q維常態分布隨機向量X=(X_1,...,X_q).假設EX_i=μ_i; VarX_i=σ_i^2, i=1,...,q, 且corr(X_j ,X_l)=ρ_jl; j=1,...,q; l=1,...,q.根據取自上述分布之樣本(X_1i,...,X_qi); i = 1,...,n,本文之目的在計算下列之假設
H_0: μ_1=...=μ_q; σ_1^2=...=σ_q^2; ρ_12=...=ρ_(q-1)q;
及
H_1: H_0 不成立;
之最大概似比檢定。;Based on a sample (X_1i,...,X_qi), i=1,...,n, obtained from a q-dimensional Gaussian distribution with EX_i=μ_i; VarX_i =σ_i^2, i=1,...,q, and corr(X_j,X_l)=ρ_jl; j=1,...,q; l=1,...,q, the purpose of this paper is to find the generalized likelihood ratio test for
H_0: μ_1=...=μ_q; σ_1^2=...=σ_q^2; ρ_12=...=ρ_(q-1)q;
versus
H_1: H_0 is not true. |
