1
Master's degree student in statistics, University of Isfahan
2
Assistant Professor, Department of Statistics, University of Isfahan
Abstract
Many multivariate control charts are based on the assumption of multivariate normality. However, the reality is that such information is rarely available and in practice, the distributional assumptions may not hold exactly. In such situations, the performance of parametric control charts is poor and unreliable. To overcome this issue, nonparametric or free distribution control charts are proposed. In this paper, we introduce a nonparametric multivariate control chart based on the two-sample Mann-Whitney test and compare its performance with the traditional Hotelling's T2 chart based on simulated data, and show that there are situations in which the nonparametric multivariate Mann-Whitney control chart has a better performance compared to the Hotelling's T2 chart.
