摘要
异常点识别是统计诊断的主要研究问题之一.本文利用均值漂移模型和似然比检验,研究了具有Rao简单结构的多元t-模型的异常点识别问题,给出了判别异常点的方法.
The identification of outlier is one of interesting issues in the statistical diagnostics.In this paper,by using the approaches of mean-shift regression model and the likelihood ratio test,the outlier identification in a multivariate t-model with Rao＇s simple structure is investigated,and the outlier detection criteria is obtained.
出处
《南华大学学报（自然科学版）》
2010年第3期75-79,共5页
Journal of University of South China：Science and Technology
关键词
极大似然估计
RAO简单结构
均值漂移模型
似然比
t-模型
maximum likelihood estimation
Rao＇s simple structure
mean-shift regression model
likelihood ratio
t-model
