两个正态总体的似然比检验
Comparison of Two Normal Distributions
-
摘要: 对于两个正态总体N(μ1,σ12)和N(μ2,σ22),本文讨论了统计假设 H0:μ1=μ,σ12=σ22 ←→ H1:μ1≠μ,σ12≠σ22。给出了似然比检验统计量的分布及相应的临界值表,并对其临界值表的计算给出了详细的讨论。Abstract: For two normal populationsN(μ1,σ12)andN(μ2,σ22), the current paper investigates the hypothesis test problem: H0:μ1=μ,σ12=σ22 versus H1:μ1≠μ,σ12≠σ22 The corresponding likelihood ratio statistic is derived and its distribution is discussed. We present the criteria table when a = 0.05 and a = 0.01. The strategy of calculating the criteria table is discussed in detail, and the corresponding SAS codes are shown also.