Maximum Likelihood Estimation of Multivariate Mixed Normal Distribution Parameters
From the perspective of hierarchy,this paper first deduces and summarizes the statistical properties of the mixed variables in the generalized multivariate mixed normal distribution formula proposed by Reinaldo B.Arellano-Vall.Secondly,combined with the derived properties of mixed variables,the EM algorithm and the covariance matrix parametric decomposition method are used to solve the problem of maximum likelihood estimation of the parameters of multivariate mixed normal under the constraints of disorder and simple tree half-order constraints.Finally,the data simulation is carried out with the help of MATLAB software,and the maximum likelihood estimation results of the covariance matrix under the half-order constraint of the simple tree are given.
multivariate mixed normal distributionEM algorithmmaximum likelihood estimationorder constraints
NETL
NSTL
万方数据
