期刊文献+

一种高维数据流的稳健监控方法

Robust monitoring method of high-dimensional data streams
下载PDF
导出
摘要 基于高维数据流在实时检测问题中的重要性,将EWMA统计量与拟合优度检验及异质混合物检验相结合,提出一种监控高维数据流的新方法。经过大量模拟和研究发现,该方法不仅实用,且监控比较稳健。 Based on the importance of high-dimensional data streams in real-time detection, in this paper, we will connect the heterogenous mixture detection, goodness-of-fit and the EWMA control chart, and then use the new method to monitor the high-dimensional data streams. Through the lots of simulations and studies, the monitoring is not only practical but also robust.
出处 《天津职业技术师范大学学报》 2016年第2期57-59,共3页 Journal of Tianjin University of Technology and Education
基金 国家自然科学基金资助项目(11271205)
关键词 高维数据流 拟合优度检验 EWMA 统计过程控制 high-dimensional data streams goodness-of-fit EWMA statistical process control
  • 相关文献

参考文献8

  • 1TARTAKOVSKY A G,VEERAVALLI V V.Asymptotically optimal quickest change detection in distribution sensor systems[J].Sequential Analysis,2008(27):441-475.
  • 2MOUSTAKIDES G V.Optimal stopping times for detecting changes in distributions[J].The Annals of Statistics,1986,14:1379-1387.
  • 3TARTAKOVSKY A G,ROZOVSKII B L,BLAZEK R B,et al.Detection of intrusions in information systems by sequential change-point methods[J].Statistical Methodology,2006,3:252-340.
  • 4MEI Y.Efficient scalable schemes for monitoring a large number of data streams[J].Biometrika,2010,97(2):419-433.
  • 5DONOHO D,JIN J.Higher criticism for detecting sparse heterogeneous mixture[J].The Annals of Statistics,2004,32(2):962-994.
  • 6ANDERSON T W,DARLING D A.Asymptotic theory of certain“goodness of fit”criteria based on stochastic processes[J].The Annals of Mathematical Statistics,1952,23(2):193-212.
  • 7ZHANG J.Powerful goodness-of-fit tests based on likelihood ratio[J].Journal of the Royal Statistical Society:Series B,2002,64(2):281-294.
  • 8ZOU C L,WANG Z J,ZI X M,et al.An efficient online monitoring method for high-dimensional data streams[J].Technometrics,2014,57(3):374-387.

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部