Multivariate Cusum Control Charts Based on data depth For Preliminary Analysis

In this paper, a CUSUM control chart based on data depth is considered for detecting a shift in either in the mean vector, the covariance matrix, or both of the process. The proposed chart is preferable from a robustness point of view, has attractive detection performances and can be useful in preliminary analysis setting where there is limited knowledge about the underlying process. A diagnostic aid is also given to estimate the location of the change.Using simulation, approximate values are given for the variance values of the plotted statistics and an upper control limit.Moreover,the new control chart can detect small sample . Unlike the existing nonparametric methods,which can detect only large sample.