Improved bootstrap X¯ control chart for non-normally distributed data

改进的非正态分布数据的自助法 X¯ 控制图

阅读:1

Abstract

This article proposes new method to improve the performance of bootstrap control chart for non-normal data. Bootstrap control charts for monitoring data require attention because the average run length (ARL) results of the bootstrap control charts can be less accurate and lacks stability. To deal with this issue, X¯ control chart from non-normal distribution was proposed using a novel improved Bootstrap method. The control limit is constructed using improved bootstrap to ensure that the UCL and LCL obtained actually give the desired ARL(0) value. Comparison of performance is studied among the Bajgier bootstrap control chart, Liu Tang Bootstrap, and their Improved bootstrap control chart. The result show that the proposed bootstrap control chart has better accuracy than its classic method. Key points:•A novel improved bootstrap method is proposed to deal with the ARL results of the bootstrap control charts that can vary.•The new X¯ Control Chart from non-normal distribution was constructed using a novel improved bootstrap method.

特别声明

1、本页面内容包含部分的内容是基于公开信息的合理引用;引用内容仅为补充信息,不代表本站立场。

2、若认为本页面引用内容涉及侵权,请及时与本站联系,我们将第一时间处理。

3、其他媒体/个人如需使用本页面原创内容,需注明“来源:[生知库]”并获得授权;使用引用内容的,需自行联系原作者获得许可。

4、投稿及合作请联系:info@biocloudy.com。