摘要
主要考虑了Cox模型的参数估计和变量选择问题,把SCAD方法引入到Cox模型中,通过BIC准则选取适当的惩罚参数,同时进行变量选择和参数估计。结合LLA(local linear approximation)逼近法和坐标下降法给出了一种有效的迭代算法。最后选取沪深A股2005～与传统逐步回归方法进行比较,结果表明基于SCAD方法的Cox模型要优于逐步回归方法建立的Cox模型。
The paper considered the parameter estimation and variable selection of the Cox model.The SCAD method was introduced into the Cox model,and the appropriate penalty parameters were selected by the BIC criterion,the variable selection and parameter estimation can be performed at the same time.An efficient iterative algorithm was proposed by combining LLA(local linear approximation)method and coordinate descent method.Finally,we selected the Shanghai and Shenzhen A shares 2005-2015 years of real estate listed companies as research samples,the prediction accuracy with the traditional stepwise regression method were compared.The results show that SCAD method based on Cox model is superior to the method of stepwise regression Cox model.
作者
郎英
苗新利
顾凯
LANG Ying;Miao Xin-Li;Gu Kai(School of Mathematics and Statistics,Chuxiong Normal University,Chuxiong Yunnan,675000)
出处
《山西大同大学学报（自然科学版）》
2019年第2期38-42,45,共6页
Journal of Shanxi Datong University(Natural Science Edition)
基金
楚雄师范学院校级一般项目[XJYB1708]
