基于卡方统计量的非参数轮廓变点识别研究[EB/OL]
北京：中国科技论文在线
） 摘要： 轮廓监控中的变点识别问题是质量控制中统计过程控制Phase I的重要研究内容
统计量值最大值是样本变异的广义似然比检验统计量来确定出方法识别的变点
同时将方法应用于京津冀地区空气质量指数的变点识别
以非参数轮廓数据为研究对象
） Abstract： Change point identification in profile monitoring is an important research topic in Phase I of statistical process control
In this paper, the profile has nonparametric characteristic and one change point
All based on the chi-square statistic, and use multivariate T2 statistics, statistical a maximum value of sample variation of generalized likelihood ratio test statistic to determine the methods to identify change point
Through a lot of simulation, method in identifying the changing point stability has excellent performance
At the same time the method is applied to air quality index of Beijing-Tianjin-Hebei region variable point recognition, recognition results can well explain time characteristics of the air quality index of Beijing-Tianjin-Hebei region
Keywords： Nonparametric profile
