earticle

영어

A sample with special missing pattern called a monotone sample or a monotone missing data pattern occurs in many applications. The statistical methods for analyzing a monotone sample have been developed by several analysts. The purpose of this paper is to introduce the maximum likelihood estimation of parameters when there is a multivariate normal sample with 2-step monotone pattern, and to present the maximum likelihood estimates of parameters in closed
forms. These estimates can be obtained by maximizing the overall likelihood function which is expressed in terms of the marginal and conditional likelihood functions with respect to parameters. The maximum likelihood estimates presented are expected to he used to develop statistical methods for analyzing a multivariate normal sample with 2-step monotone pattern in applications.