Author： Hamdan Hasan
Publisher： Taylor & Francis Ltd
ISSN： 0361-0926
Source： Communications in Statistics: Theory and Methods, Vol.35, Iss.3, 2006-01, pp. : 407-413
Disclaimer: Any content in publications that violate the sovereignty, the constitution or regulations of the PRC is not accepted or approved by CNPIEC.
Abstract
The purpose of this study is to approximate and identify infinite scale mixtures of normals, SMN. A new method for approximating any infinite SMN with a known mixing measure by a finite SMN is presented. In the new method, the modulus of continuity of the normal family as a function of the scale is used to discretize the mixing measure. This method will be used to approximate univariate and multivariate SMN with mean 0. In the multivariate case, two different methods are used to approximate the infinite SMN. Several results related to SMN are proved and other known ones are presented. For example, SMN are characterized by their corresponding Laplace transforms.
Related content
Access to resources
Recommend
A new sufficient condition for identifiability of countably infinite mixtures
Metrika, Vol. 77, Iss. 3, 2014-04 ,pp. :