Investigation of the Multiple Imputation Method in Different Missing Ratios and Sample Sizes

In many studies, missing data are the real trouble to researchers. Because the statistical methods are designed for complete data sets. Multiple imputation method is developed to solve the missing data problem. The method is also used effectively in some useful properties of the Bayes method. If there are missing values in the data set, Bayesian method can be used to prevent the loss of information. In this study, the performance of the multiple imputation method is evaluated by generating survival data with different missing rates and different sample sizes. Also, informative priors and multiple imputation method are used together to prevent the missing information in the variable with missing value.