Detection of outlier in circular data and single circular regression have received much attention. Several diagrammatical plots and numerical presentation have been proposed to identify the existence of outliers. Currently, outlier detection analysis in multiple circular regression model is also attracting the interest of statisticians and researchers to do the research in depth. In this paper, the outlier detection methods on multivariate eye data in the multiple circular regression model is considered. The model properties of multiple independent circular variable are presented. Five statistics for outlier detection methods has been investigated and compared using graphical and numerical methods. The multivariate eye data set is considered in this study to investigate the performance of the proposed statistics. It is found that the proposed statistics are good in identifying outliers not limited to the presence of a single or few outlier, but able to identify the presence of multiple outliers at the same time.
