Advances and Applications in Statistics
Volume 60, Issue 1, Pages 11 - 33
(January 2020) http://dx.doi.org/10.17654/AS060010011 |
|
ROBUST ALTERNATIVES TO MEWMA CONTROL CHART USING DISTANCE-BASED AND COORDINATE-WISE ROBUST ESTIMATORS
J. Faridzah, H. A. Hazlina and S. Y. Sharipah Soaad
|
Abstract: Control chart is one of the primary techniques used in statistical process control. In Phase I, historical data are used to construct the chart and to determine whether the process has been in control over the period of time when the data were collected. However, the presence of multiple outliers could mar the performance of the control charts if go undetected. Due to the masking and swamping effects of outliers, control charts such as MEWMA E2 could be in jeopardy. To mitigate the issue of outliers, this paper proposes robust alternatives to MEWMA E2 control chart which are based on distance-based and coordinate-wise robust estimators. Simulation studies show that the proposed control charts are more robust than the conventional MEWMA E2 especially when the percentage of outliers is high regardless of sample sizes and number of dimensions. |
Keywords and phrases: MEWMA control chart, minimum vector variance, modified one-step M-estimator, false alarm, probability of detection.
|
|
Number of Downloads: 327 | Number of Views: 907 |
|