中文摘要：

图模型是处理高维数据的有力工具，其中图模型的选择是统计推断的重要方面．利用E—MS算法进行缺失数据的图模型选择．E—MS算法是基于EM迭代思想的，把模型下的参数和模型选择参数统一为新的参数，这样模型选择就成为E—M迭代过程的一部分．最后给出了3，4，5变量情形下的模拟研究。

英文摘要：

Graphical model is a powerful tool to deal with high dimensional data, and the graphical modal selection is an important aspect of statistical inference. In this paper, the E - MS algorithm was used to select the appropriate graphical model with missing data. The E - MS algorithm is based on the idea of EM iteration, and combines the parameters of the model with the model selection parameters as the new parameters, so the model selection becomes the part of the E - M iterative process. Finally, a simulation study of 3,4 and 5 variables was given.