MODELLING STOCK RETURNS USING A NORMAL VARIANCE MEAN MIXTURE WITH FINITE MIXTURE OF WEIGHTED INVERSE GAUSSIAN DISTRIBUTIONS

Abstract

The Normal Weighted Inverse Gaussian distribution arises as a Normal Variance Mean mixture with Weighted Inverse Gaussian mixing distribution. Appropriate choice of the weight for Inverse Gaussian gives rise to finite mixture of special cases of Generalized Inverse Gaussian (GIG) distributions. This article deals with a kind of weight which lead to a finite mixture between a GIG distribution of indexes