• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文会议>31st annual national conference of the American Society for Engineering Management 2010 >NONPARAMETRIC CHANGE-POINT ANALYSIS FOR SLOPE VARIABILITY
【24h】

NONPARAMETRIC CHANGE-POINT ANALYSIS FOR SLOPE VARIABILITY

机译：边坡变化的非参数变化点分析

获取原文
获取原文并翻译 | 示例
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

Living systems present non-normal behavior and trends of growth or decrement. To manage and improve, control charts are the obvious tool to understand these patterns, and be able to detect whether they are in statistical control. However, most charts, even the ones designed to detect changes in trends, assume homoscedasticity through time, and the use of additional control charts to monitor the spread helps to validate this assumption. Nevertheless, the time a chart for variance signals an out of control point does not imply that the change initiated at that moment. This problem increases in the presence of small and sustained shifts. An approach designed to estimate the right moment of change is the change-point analysis technique. This paper proposes a change-point model that uses a clustering technique and the Median Test to estimate the moment of a change in variance, in conditions of non-normal behavior and defined trends.
机译：生命系统呈现非正常行为以及增长或减少的趋势。为了进行管理和改进，控制图是了解这些模式并能够检测它们是否处于统计控制中的显而易见的工具。但是，大多数图表，甚至那些旨在检测趋势变化的图表，都假设其随时间推移具有均方差性，并且使用其他控制图来监控价差有助于验证这一假设。但是，方差图发出信号表示超出控制点的时间并不意味着该时刻开始更改。在出现小而持续的班次时，这个问题会增加。一种用于估计正确的变更时刻的方法是变更点分析技术。本文提出了一个变更点模型，该模型使用聚类技术和中值检验来估计在非正常行为和已定义趋势的情况下方差变化的时刻。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱：kefu@zhangqiaokeyan.com

京公网安备：11010802029741号 ICP备案号：京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有

  • 客服微信

  • 服务号