Multivariate control chart based on robust estimator
A Hotelling T 2 control chart has been widely used for monitoring first phase of multivariate statistical process control. However, this classical control chart which is based on classical estimators is not effective in detecting multivariate outliers. As an alternative, T 2 statistics based on r...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Published: |
Penerbit ukm
2009
|
| Online Access: | http://journalarticle.ukm.my/1901/ http://journalarticle.ukm.my/1901/ |
| Summary: | A Hotelling T 2 control chart has been widely used for monitoring first phase of multivariate
statistical process control. However, this classical control chart which is based on classical
estimators is not effective in detecting multivariate outliers. As an alternative, T 2 statistics
based on robust minimum volume ellipsoid or the minimum covariance determinant are
proposed. The performance of these robust control charts are investigated extensively by real
examples and Monte Carlo simulation. The result indicates that the robust control charts is
more effective in detecting outliers than the classical control charts |
|---|