School of Mathematical and Statistical Sciences Faculty Publications and Presentations

A Robust Bootstrap Control Chart for the Log-Logistic Percentiles

Document Type

Article

Publication Date

1-22-2022

Abstract

In this paper, we develop a robust bootstrap control chart for the problems of detecting a shift in the percentile of the log-logistic population in a process monitoring scenario, in which the quality characteristic of interest is product lifetime. The construction of a quality control chart usually depends on an accurate estimation of the unknown process parameters, which is often obtained based on conventional methods such as the maximum likelihood and methods-of-moments estimators. However, these estimators are sensitive to outliers and could result in severe bias for obtaining the control limits and thus misleading false alarm signals. To overcome this potential issue, we advocate a robust repeated median estimator as an alternative for the process parameters. Simulation studies and a real-data application are provided to illustrate the effectiveness of the proposed bootstrap control charts in terms of the average run length and the standard deviation of run length.

Comments

Copyright © 2022, Grace Scientific Publishing

https://rdcu.be/dYAFJ

Publication Title

Journal of Statistical Theory and Practice

DOI

https://doi.org/10.1007/s42519-021-00239-3

Link to Full Text

Share

COinS