Robust Fitting of a Wrapped Normal Model to Multivariate Circular Data and Outlier Detection

In this work, we deal with a robust fitting of wrapped normal model to multivariate circular data. Robust estimation is supposed mitigate the adverse effects outliers on inference. Furthermore, use proper method leads definition effective outlier detection rules. achieved by suitable modification classification-expectation-maximization algorithm that has been developed perform maximum likelihood parameters distribution. The concerns complete-data estimating equations involve set data dependent weights aimed downweight effect possible outliers. Several techniques are considered define weights. finite sample behavior resulting proposed methods investigated some numerical studies and real examples.