Modified minimum covariance determinant estimator and its application to outlier detection of chemical process data

To overcome the main flaw of minimum covariance determinant (MCD) estimator, i.e. difficulty to determine its main parameter <italic>h</italic>, a modified-MCD (M-MCD) algorithm is proposed. In M-MCD, the self-adaptive iteration is proposed to minimize the deflection between the standard deviation of robust mahalanobis distance square, which is calculated by MCD with the parameter <italic>h</italic> based on the sample, and the standard deviation of theoretical mahalanobis distance square by adjusting the parameter <italic>h</italic> of MCD. Thus, the optimal parameter <italic>h</italic> of M-MCD is determined when the minimum deflection is obtained. The results of convergence analysis demonstrate that M-MCD has good convergence property. Further, M-MCD and MCD were applied to detect outliers for two typical data and chemical process data, respectively. The results show that M-MCD can get the optimal parameter <italic>h</italic> by using the self-adaptive iteration and thus its performances of outlier detection are better than MCD.