基于Markov Chain Monte Carlo模型对医院出院病人调查表数据缺失的填补与分析

目的对医院出院病人调查表普遍存在的数据缺失进行填补与分析,以保证统计调查表的质量,为医院以及上级卫生部门了解现状,进行预策和决策提供技术支持和质量保证。方法运用SAS9.1,采用多重填补方法Markov Chain Monte Carlo(MCMC)模型对缺失数据进行多次填补并综合分析。结果MCMC填补10次的结果最优。结论(Multiple Imputation)MI方法在解决医院出院病人调查表数据缺失时有优势,发挥空间较大,且填补效率较高。

The Imputation and Analysis for the Missing Data of Survey of Patients Discharged from Hospital

Objective To approach the imputation and examination methods to solve the data missing problem generally existed in the survey of patients discharged from hospital,so that the statistic forms data quality and the skill support for examination and decision in the hospital and the upper health department knowing the present situation can be guaranteed.Methods Using SAS9.1,to fill up multiply and analyze comprehensively the missing data with Markov Chain Monte Carlo(MCMC) model.Results MCMC method is the best after 10 times imputations.Conclusion MI method has advantages solving the data missing of the survey of patients discharged from hospital,and it also has larger space and higher efficiency imputation.