This paper is devoted to providing an iterative algorithm for computing the maximum likelihood estimators of discrete distributions under an increasing convex order constraint,and the convergence of the algorithm is proved.

To keep the advantage, this thesis brings forward ameliorated model based on computing model of maximum likelihood estimation and revises centroid algorithm value of coordinates and then optimizes the model based on centroid algorithm.

This disquisition revises the error of node localization based on measuring distance by maximum likelihood estimation which can calculate the error between surveying distance and computing distance to enhance precision of node orientation and keep the advantage of its smiple algorithm.

A non-parametric maximum likelihood estimation method for estimating infection curve of infectious diseases;

In this paper,we introduce the essence of the maximum likelihood parameter estimation,the EM algorithm used for mixture model parameter estimation,and give a computer emulation example to illustrate validity and convergence of EM algorithm.

When errors is a AR(1) time series,we studied the quasi-likelihood equation for the semiparametric model,and investigated the existence of quasi-maximum likelihood estimators.