Missing data is a common issue in clinical trials with increasing attention from regulatory agencies. It's widely accepted that there's no universal solution to missing data. The new guideline and recommendation is to specify a primary approach to handle missing data, and construct a range of sensitivity analyses under different assumptions to assess the robustness of the primary assumption about missing data. We performed a case study for continuous endpoint with variety of missing imputation methods in practice using SAS procedures and macros, and compared the results. The imputation methods discussed including single imputation methods, multiple imputation, mixed models, and common pattern mixture model methods under MNAR assumption.