论文部分内容阅读
目前，回归诊断不仅用于一般线性模型的诊断，还被逐步推广应用于广义线性模型领域（如用于Ｌｏｇｉｓｔｉｃ回归模型），但由于一般线性模型与广义线性模型在残差分布的假定等方面有所不同，所以推广和应用还存在许多问题。鉴于此，作者提出一种新的图示方法——经验Ｐｅａｒｓｏｎ残差图，用以考察Ｌｏｇｉｓｔｉｃ模型的拟合优度及检测异常点．
At present, regression diagnosis is not only used for the diagnosis of general linear models, but has also been gradually applied to the generalized linear model domain (such as for Logistic regression model), but due to the general linear model and the generalized linear model in the assumption of residual distribution, etc. Different, so there are still many problems in promotion and application. In view of this, the author proposes a new graphical method, the empirical Pearson residual graph, to investigate the goodness of fit of the Logistic model and detect abnormal points.