摘要
在许多情况中,我们需要对logsitic回归模型中的重要变量进行选择,这适用于预测问题,本文从信息论准则出发提出了一种选择程序,并且证明了这些程序是强相合的。
In many situation, we are interested in selection of important variables which are adequate for prediction under a logistic regression model. In this paper, some selection procedures based on the information theoretic criteria are proposed, and these procedures are proved to be strongly consistent.
基金
The reseach is supported by the U.S. Air Force of Scientific Resseach under Contract F49520-85-C-0008. These authors are also supported by National Natural Science Foundation of China.
关键词
LOGISTIC回归
变量选择
相合性
Consistency, information theordtic criterion, logistic discrimination, logistic regression, maximum likelihood, model selection.
