变量选择方法在医疗保险赔付评估中的应用
|
摘要
在广义线性模型假设下,采用Lin的医疗费用模型,运用LASSO和SCAD方法对影响医疗费用的因素进行选择,并对两种方法的有效性进行了对比分析,从而得出影响医疗保险赔付的重要因素,解决了高维变量带来的一系列问题。实例分析中,由于两种方法注重的统计性质不同,选择出的解释变量略微不同,但通过分析发现,两种结果都具有良好的解释性,反映了影响医疗保险赔付的重要信息。
Under the assumption of the general linear model,variable selection method is applied to study the factors which have more significant effects in compensation for medical insurance.First of all,LASSO and SCAD methods are employed to select the important variables,and a comparison is given to see their effects based on the different methods,and hence then,it gives the important effect factors in compensation for medical insurance.Some simulation studies with the analysis of the real data of medical insurance are preformed to assess the proposed model and the methods in this paper.
Under the assumption of the general linear model,variable selection method is applied to study the factors which have more significant effects in compensation for medical insurance.First of all,LASSO and SCAD methods are employed to select the important variables,and a comparison is given to see their effects based on the different methods,and hence then,it gives the important effect factors in compensation for medical insurance.Some simulation studies with the analysis of the real data of medical insurance are preformed to assess the proposed model and the methods in this paper.
引文
[1]杨馥,刘珺.“十二五”期间陕西人身保险市场需求变化研究[J].西安财经学院学报,2012(1).
[2]Moran J L,Solomon P J,Aaron P R.New Models for Old Questions:Generalized Linear Models for Cost Prediction[J].Journal of Evaluation in Clinical Practice,2007(3).
[3]仇春娟,陈滔.商业医疗保险损失分析:基于广义线性模型的实证研究[J].应用概率统计,2012(4).
[4]赵晓兵,王伟伟.高维附加信息下的商业医疗保险费用评估模型和方法[J].财经论丛,2013(4).
[5]Lin D Y.Regression Analysis of Incomplete Medical Cost Data[J].Statistics in Medicine,2003(7).
[6]Tibshirani R.Regression Shrinkage and Selection via the Lasso[J].Journal of the Royal Statistical Society,1996(1).
[7]Zou Hui.The Adaptive Lasso and Its Oracle Properties[J].Journal of the american Statistical Association,2006(476).
[8]Fan Jianqing,Li Runze.Variable Selectionvia Nonconcave Penalized Likelihood And Its Oracle Properties[J].Journal of the American Statistical Association,2001(456).
[9]Zhang Riquan,Zhao Weihua,Liu Jicai.Robust Estimation and Variable Selection for Semiparametric Partially Linear Varying Coefficient Model Based on Modal Regression[J].Journal of Nonparametric Statistics,2013(2).
[10]Zou Hui,Li Runze.One-Step Sparse Estimates in Nonconcave Penalized Likelihood Models[J].The Annals of Statistics,2008(4).
[2]Moran J L,Solomon P J,Aaron P R.New Models for Old Questions:Generalized Linear Models for Cost Prediction[J].Journal of Evaluation in Clinical Practice,2007(3).
[3]仇春娟,陈滔.商业医疗保险损失分析:基于广义线性模型的实证研究[J].应用概率统计,2012(4).
[4]赵晓兵,王伟伟.高维附加信息下的商业医疗保险费用评估模型和方法[J].财经论丛,2013(4).
[5]Lin D Y.Regression Analysis of Incomplete Medical Cost Data[J].Statistics in Medicine,2003(7).
[6]Tibshirani R.Regression Shrinkage and Selection via the Lasso[J].Journal of the Royal Statistical Society,1996(1).
[7]Zou Hui.The Adaptive Lasso and Its Oracle Properties[J].Journal of the american Statistical Association,2006(476).
[8]Fan Jianqing,Li Runze.Variable Selectionvia Nonconcave Penalized Likelihood And Its Oracle Properties[J].Journal of the American Statistical Association,2001(456).
[9]Zhang Riquan,Zhao Weihua,Liu Jicai.Robust Estimation and Variable Selection for Semiparametric Partially Linear Varying Coefficient Model Based on Modal Regression[J].Journal of Nonparametric Statistics,2013(2).
[10]Zou Hui,Li Runze.One-Step Sparse Estimates in Nonconcave Penalized Likelihood Models[J].The Annals of Statistics,2008(4).