CMStatistics 2018: Start Registration
View Submission - CMStatistics
Title: A wrapped skew generalised normal family Authors:  Theodor Loots - University of Pretoria (South Africa) [presenting]
Andriette Bekker - University of Pretoria (South Africa)
Abstract: A scale mixture of the uniform and generalised gamma distribution results in a very flexible parametric family of distributions. This family includes a skew-generalised normal distribution as a special case. The resultant family is wrapped in the usual fashion, yielding a new wrapped parametric family of distributions. Simulations are carried out using the corresponding stochastic representation, while parameter estimation is performed using the method of trigonometric moments, as well as that of maximum likelihood. Finally, this new family is fitted to natural phenomena such as wind directional data.