Statistical Inference for Stable Distribution Using EM algorithm

2018/11/12 by Teimouri, Mahdi
#FOS: Mathematics #Statistics Theory (math.ST)

paper · doi:10.48550/arxiv.1811.04565

Abstract

The class of α-stable distributions with a wide range of applications in economics, telecommunications, biology, applied, and theoretical physics. This is due to the fact that it possesses both the skewness and heavy tails. Since α-stable distribution suffers from a closed-form expression for density function, finding efficient estimators for its parameters has attracted a great deal of attention in the literature. Here, we propose some EM algorithm to estimate the maximum likelihood estimators of the parameters of α-stable distribution. The performance of the proposed EM algorithm is demonstrated via comparison study in the presence of other well-known competitors and analyzing three sets of real data.

Related

• Statistical Inference for Mixture of Cauchy Distributions
• Likelihood-free Bayesian inference for alpha-stable models
• Alpha Estimation via Sample Splitting: A Two-Sample Framework for Stable-like Distributions
• Estimation of stability index for symmetric α-stable distribution using quantile conditional variance ratios
• alphastable: An R Package for Modelling Multivariate Stable and Mixture of Symmetric Stable Distributions