Mcar Missing Mechanism Based On The Location Parameters Of The Linear Model Estimation Problem

Under the special mechanism, we attempt to get the mean estimation, the variance estimation and the best method closer to the real value. It becomes the most popular subject of dealing with the missing data problem and it is the core of this article. Firstly, the article presents some basic knowledge about missing data, simply introduces some important methods about the missing data problem and the process of the development of the missing data. In this article, we consider the situation as follows:under the MCAR mechanism, we use the information of the multidimensional auxiliary variables to establish a multiple linear model, and then use the Least Squared method to estimate the regression coefficients. Finally, we get the parametric estimations. In the third chapter, we argue about the mean estimation and the Monte Carlo simulation proves the advantage and efficiency of the positional parametric. In the third chapter, we argue about the variance estimation. In this chapter, we introduce the variance estimation given by the standard formula, the Jackknife method and the improved Jackknife method. At the last part of this chapter, we also use the Monte Carlo simulation to validate the advantage of the Jackknife estimation.