摘要
系统对含定性变量的回归模型进行比较研究,通过引入虚拟变量处理自变量含定性变量的回归问题与折线回归,探讨了处理因变量含定性变量的线性概率模型、Probit模型与Logistic模型的区别与联系,以便针对具体情况选择合适的模型,最后利用SPSS自带数据给出实例分析.
This paper makes a comparative study of regression models with qualitative variables. By introducing virtual variables to deal with the regression problems of independent variables with qualitative variables and polyline regression,the differences and relations between linear probability model,Probit model and Logistic model for dealing with dependent variables with qualitative variables are discussed,so that readers can choose appropriate models for specific situations. Finally,an example is given by using SPSS data.
作者
王丙参
陈庆美
魏艳华
WANG Bingcan;CHEN Qingmei;WEI Yanhua(School of Statistics,Capital University of Economics and Business,Beijing 100070;School of Mathematics and Statistics,Tianshui Normal University,Tianshui Gansu 741001)
出处
《宁夏师范学院学报》
2019年第10期21-26,共6页
Journal of Ningxia Normal University
基金
国家自然科学研究基金(11561060)