基于经验欧氏似然的均值单变点检测  

On Mean Change-Point Detection Based on Empirical Euclidean Likelihood

作　　者：张军舰 李智航 ZHANG Junjian;LI Zhihang(College of Mathematics and Statistics,Guangxi Normal University,Guilin,541004,China)

机构地区：[1]广西师范大学数学与统计学院,桂林541004

出　　处：《应用概率统计》2021年第1期47-58,共12页Chinese Journal of Applied Probability and Statistics

基　　金：国家自然科学基金项目(批准号:11861017)资助。

摘　　要：均值单变点检测是研究变点问题的基础.论文根据均值单变点模型的特点,构造截断经验欧氏似然比检验函数并给出显式表达.在此基础上,得到了零假设下检验统计量的极限分布为极值分布,给出变点的诊断方法.在有变点的情况下,进一步给出变点位置的估计和其相合性的理论证明.最后通过数值模拟和尼罗河年流量的实证分析说明所提方法的有效性和实用性.Mean chage-point detection is the groundwork in statistics.The trimmed empirical Euclidean likelihood ratio function is constructed based on the features of change-point in mean model.And the explicit expression is derived.The null limit distribution of the test statistic is investigated with extreme value distribution.And the change-point detection is made.if the change-point exists,it’s location and consistency are discussed.Simulations and real analysis of Nile River data show that our proposed method is practicable and effective.

关 键 词：均值变点 经验欧氏似然 截断 极值分布 

分 类 号：O212.7[理学—概率论与数理统计]