Parameter Estimation of the Extended Generalized Gaussian Family Distributions using Maximum Likelihood Scheme

摘要

An extended generalized Gaussian distribution which can describe a family of symmetric and asymmetric distributions is considered. Parameter estimation of this function using maximum likelihood scheme is proposed. By measured the tail length and skewness of the observed data, the method integrates a pre-calculated table of initial values for parameters estimation. This allows a fast convergence of the presented model for real-time applications. The simulation results also show that the proposed scheme is an asymptotically unbiased estimator in terms of Cramer- Rao lower bound criterion.