首页 | 本学科首页   官方微博 | 高级检索  
     检索      

线性模型信息准则的稳健性
引用本文:黄荣坦.线性模型信息准则的稳健性[J].厦门大学学报(自然科学版),1993,32(5):563-567.
作者姓名:黄荣坦
作者单位:厦门大学数学系
摘    要:在不需要假定:1)被选模型包含真实模型;2)误差分布为正态分布的条件下,应用作者定义的广义K-L差异度,得到了自变量选择的广义K-L差异度准则。这个准则包含了所有信息准则作为它的特例,由此说明信息准则具有稳健性。

关 键 词:线性回归模型  信息准则  鲁棒性

Robustness of Information Criteria for Variable Selection in Linear Regression Model
Huang Rongtan.Robustness of Information Criteria for Variable Selection in Linear Regression Model[J].Journal of Xiamen University(Natural Science),1993,32(5):563-567.
Authors:Huang Rongtan
Institution:Dept. of Math.
Abstract:In references,the conduction of infomation criteria for variable selectionin linear regression model needed some strict conditions. In practice,the effection of the use of these criteria doesn' t become very bad when these strict conditions are violated. This fact tells us that the infomation criteria are robust, The infomation criteria obtained under certain mild conditions. These mild conditions mean the robustness of infomation criteria.
Keywords:Linear regression model  Variable selection criteria  Infomation criteria  General K-L discrepancy  General K-L discrepancy criteria  Robustness  
本文献已被 CNKI 维普 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号