Based on kernel function and non-parametric multiple imputation algorithm to solve the problem of missing data

To fill the data missed and to make statistical activities more effectively response, according to the missing mechanism the data is missing at random (MAR) and missing completely at random (MCAR) conditions, several methods for missing data, multiple imputation, kernel function of non-parametric regression analysis are introduced. The kernel function of non-parametric regression analysis is analyzed with emphasis, and the corresponding functions and the theorem for their analytical methods are discussed. Then, an example is introduced to verify the accuracy of this method. By analyzing, the results show that the data filled according to the methods of kernel function of non-parametric regression analysis can replace the original data, and the statistical results are identified with the original ones.