Volume 16, Number 2 (2017) | JIRSS 2017, 16(2): 0-0 | Back to browse issues page

XML Print

Abstract:   (349 Views)

This paper presents a new mixture model via considering the univariate skew Laplace distribution. The new model can handle both heavy tails and skewness and is multimodal. Describing some properties of the proposed model, we present a feasible EM algorithm for iteratively computing maximum likelihood estimates. We also derive the observed information matrix for obtaining the asymptotic standard errors of parameter estimates. The finite sample properties of the obtained estimators together with the consistency of the associated standard error of parameter estimates are investigated by a simulation study. We also demonstrate the flexibility and usefulness of the new model by analyzing real data example.  
Type of Study: Original Paper | Subject: 60Exx: Distribution theory
Received: 2017/01/1 | Accepted: 2017/04/22 | Published: 2017/04/22