A Survey of Strategies for Modeling Cross-Classifications Having Ordinal Variables
综述了包含有序变量的交叉分类表的主要建模策略，重点介绍对数线性模型和对数比数模型的构建、解释及其应用，适用于处理有序分类数据的统计建模。
Abstract Abstract A survey is given of the main strategies for modeling cross-classifications that contain ordinal variables. Models are described for two-way tables in which one or both classifications are ordered and for multidimensional tables in which at least one classification is ordered. Primary emphasis is given to construction, interpretation, and implications of loglinear and logit models. Key Words: Loglinear modelsLogit modelsOdds ratiosOrdered categorical dataAssociationTests of independence