CFE 2019: Start Registration
View Submission - CMStatistics
Title: Circular mean variance mixture models Authors:  Priyanka Nagar - University of Pretoria (South Africa) [presenting]
Andriette Bekker - University of Pretoria (South Africa)
Mohammad Arashi - Ferdowsi University of Mashhad (Iran)
Abstract: One of the widely used methods for introducing asymmetry into a model is by means of a mean mixture approach. Many wrapped circular models have been proposed based on this method. However, the need for flexible circular models is still prevalent. A new general class of flexible spherical models is introduced based on mean variance mixture modeling. Special cases of this new class are studied in detail. Given the complex structure of this class, an EM algorithm based approach for performing maximum likelihood estimation is considered. The practicality of the proposed distribution is illustrated through a data application.