Comparison between the multiple imputation and the multiple imputation then deletion methods in treating missing values for the variables of multiple linear regression model

Abstract

process of collecting data for search variables in some cases faces the loss of some values that variables, due to several reasons, some of them is outside the will of the , others reasons is intentional, Thus, the data are described as incomplete and their reliability in the analysis leads to inaccurate results, Which calls for estimating the value of missing observations in the search variables and then adopting the complete data in the estimation process. The study dealt with two methods of incomplete data processing for regression model (Multiple Imputation)MI) and multiple imputation, then deletion (MID) (.

The aim of the research is to compare these two methods in an attempt to arrive at the best method based on simulations , The results of the simulation experiments and using the mean error squares of the estimated regression model, as a criterion for differentiation, showed that the multiple estimation method then deletion were better than the multiple estimation method in incomplete data processing.