A Graphical Analysis of Deviance
提出一种图形方法，展示广义线性模型拟合序列中所有单解释变量比较的结果，帮助识别重要变量并评估模型变量间的非正交性。
A graphical representation of all single explanatory variable comparisons in a sequence of generalized linear model fits is given and applied to some sets of data. The graph is useful in determining which are the important explanatory variables and for assessing any non‐orthogonality between the variables included in the models.