预测我国医护比例的模型研究  

Establishing models on forecasting the medical care ratio in China

作　　者：王爱青[1] 刘卫星[1] 刘晓坤 王金艳 张庆华 

机构地区：[1]唐山市中医医院,河北唐山063000 [2]唐山市工人医院老年科,河北唐山063000

出　　处：《现代医学》2017年第9期1229-1234,共6页Modern Medical Journal

基　　金：河北省卫生厅医学科学研究课题(20130318);唐山市科技项目(2014058X)

摘　　要：目的:应用残差自回归模型预测我国医护比例的变化趋势,并将其预测效能同灰色模型[GM(1,1)]的预测效能进行比较,从而为了解我国医护比例变化趋势提供依据。方法:收集1980~2014年我国医护比例的资料,用SAS 9.3统计软件构建灰色GM(1,1)模型,用EViews 8.0统计软件构建残差自回归模型,并使用建立好的统计模型对我国2015~2025年医护比例的变化趋势进行预测。结果:残差自回归模型拟合及预测的MRD、MSE、RMSE和MAE分别为2.337 8、0.000 4、0.019 0、0.014 4、2.746 8、0.000 7、0.026 8、0.026 8。GM(1,1)模型拟合及预测的MRD、MSE、RMSE和MAE分别为4.203 1、0.001 2、0.034 3、0.026 4、9.653 4、0.009 0、0.094 9、0.094 3,并使用残差自回归模型对2015~2025年我国医护比例进行了预测。结论:残差自回归模型对我国医护比例的拟合精度较高,预测效果可靠;从2014年开始我国医护比例基本持平,并呈不断上升的趋势。Objective： To establish the models of GM（1,1） and auto-regressive prediction used to predict medical care ratios in China and to compare the predictive effect among them,so as to provide the basis for grasping variation tendency of medical care ratios in China.Methods： The data on the medical care ratios from 1980 to 2014 in China were collected,SAS 9.3 and EViews 8.0 were employed to fit corresponding models,then the medical care ratios from 2015 to 2025 in China were predicted by trend extrapolation model-establishing.Results：The MRD,MSE,RMSE and MAE fitted and predicted by auto-regressive model were 2.337 8,0.000 4,0.019 0,0.014 4 and 2.746 8,0.000 7,0.026 8,0.026 8,respectively.The MRD,MSE,RMSE and MAE fitted and predicted by GM（1,1） model were 4.203 1,0.001 2,0.034 3,0.026 4 and 9.653 4,0.009 0,0.094 9,0.094 3,respectively.Afterwards the auto-regressive model was used to forecast the medical care ratio from 2015 to 2025 in China.Conclusion： Auto-regressive model is a forecasting model with higher accuracy and its forecasting effect is reliable; Since 2014 the medical care ratio begins to be equal and it is in a rising trend.

关 键 词：GM(1 1)模型 残差自回归模型 医护比例 预测 

分 类 号：R195.1[医药卫生—卫生统计学] R19[医药卫生—卫生事业管理]