我国城市消费者寿险购买行为的影响因素及预测  

The Development of Forecasting Model on Consumers Life Insurance Purchasing by Discriminant Analysis and Logistic Regression

作　　者：张阔[1] 李桂华[2] 李燕飞[1] 

机构地区：[1]南开大学政府管理学院,天津300071 [2]南开大学商学院,天津300071

出　　处：《数理统计与管理》2011年第2期291-298,共8页Journal of Applied Statistics and Management

基　　金：国家自然科学基金重点项目(No:70832001);国家自然科学基金项目(No:70572084)

摘　　要：通过文献回顾、专家访谈和问卷调查,确定了8种对消费者寿险购买行为有重要影响的个体态度变量。根据在全国10个城市进行问卷调查所获得的态度变量和人口统计学变量数据,采用判别分析和Logistic回归的方法分别建立了消费者寿险购买行为的预测模型。对模型的评价表明,2个模型都有较好的预测精度;若将两个模型的联合应用,能取得更好的预测效果,对寿险营销管理具有重要的参考价值。Through experts interview and open questionnaire investigation,eight attitude variables of consumers related to life insurance purchasing were identified.The data of 8 attitude variables and 4 demographic variables of 4694 consumers of 10 cities were achieved by sampling survey in 2007.Two predicting models on consumers life insurance purchasing were developed by discriminant analysis and logistic regression.The validity of the discriminant model and the regression model were evaluated,and the results shows they have favorable precision and are helpful to the marketing administration of life insurance.

关 键 词：消费者 寿险 预测 判别分析 LOGISTIC回归 

分 类 号：F84[经济管理—保险] O212[理学—概率论与数理统计]