A comparative study of outlier detection procedures in multiple circular regression

This paper considers outlier detection methods in multiple circular regression. It focuses on the detection of outliers in a multiple circular regression model, where both variables are circular. This model has many interesting properties related to multiple linear regression models. The graphical techniques are investigated to compare the results with the numerical method, where the extension of the COVRATIO statistic in linear regression model is applied to the circular case. In general, both methods are perform well in detecting the outliers in the data set.