摘要
原假设（ｙ１，ｙ２．．．，ｙｎ）是正态独立随机量时间序列，其均值和方差分别为μ和σ＾２备选假设为均值μ在某一时刻（未知）发生变化，本文对σ＾２为已知的情况，导出了由Ｈａｗｋｉｎｓ（１９７７）提出的似然经检验统计量Ｕ的简明且便于计算的分布函数表达式，并建立了分布函数的数值表。
An alternative to hypothesis that the sequence {y1, y2,…,yn} are independent and identically distributed normal random variables, with mean μ and variance σ2 , is that the location parameter μ shifts at some unknown instant. In this paper, a clear expression of the null distribution function of the likelihood ratio test statistics U that is considered by Hawkins (1977) is derived when σ2 is known. The numerical tables of the distribution funtion are given.
关键词
似然比检验
统计量
分布函数
Location Shift, Change-Point, Likelihood Test Statistics.
