Monitoring Location: A Nonparametric Control Chart Based on the Signed-Rank Statistic
Marien A. Graham1, Subhabrata Chakraborti2, Schalk W. Human1
1Department of Statistics, University of Pretoria, Pretoria, Gauteng, South Africa; 2Department of Information Systems, Statistics and Management Science, University of Alabama, Tuscaloosa, United States

Nonparametric control charts are useful when there is limited or complete lack of knowledge about the underlying distribution. In this talk an in-depth study of the characteristics and performance of the two-sided nonparametric exponentially weighted moving average (EWMA) signed-rank control chart will be discussed; this control chart combines the philosophies of the well-known EWMA chart and the Wilcoxon Singed-Rank (SR) statistic and was originally proposed by Amin and Searcy (1991). An important advantage of the nonparametric EWMA signed-rank chart is its inherent in-control robustness. In fact, the in-control run-length distribution and hence all of its associated characteristics (e.g., false alarm rate, average, standard deviation, median, etc.) of the chart remain the same for all unknown continuous distributions. An additional benefit is that the nonparametric EWMA signed-rank chart is more resistant to outliers than its parametric counterparts such as the EWMA for subgroup averages. An exact Markov chain approach is used to determine the exact run-length distribution and the associated performance characteristics. In order to aid practical implementation, tables are provided for the chart's design parameters. An extensive simulation study shows that on the basis of minimal required assumptions, robustness of the in-control run-length distribution and out-of-control performance, the proposed nonparametric EWMA signed-rank chart is a strong contender in many applications where traditional parametric charts are currently used.

Keywords: Distribution-free; Markov chain; Median; Robust

Biography: Mrs. MA Graham is currently enrolled for her PhD in Mathematical Statistics at the University of Pretoria, South Africa. She has been a full-time lecturer at the Department of Statistics at the University of Pretoria since 2005. Her PhD research focuses on “Modern developments and advances in statistical control charts: parametric and nonparametric.” Her supervisor is Prof. Dr. S. Chakraborti from the University of Alabama, USA and her co-supervisor is Dr. S.W. Human from the University of Pretoria. While enrolled for her Master's degree she visited Prof. Dr. S. Chakraborti at the University of Alabama, USA, twice. She was invited to contribute to the Encyclopedia of Statistics in Quality and Reliability published by John Wiley & Sons in 2007. She has also published peer reviewed articles in international journals which include Quality Engineering and Journal of Applied Statistics. Her research interests include Statistical Process Control, Quality Control and Nonparametric Statistics.