Testing for interaction in an IxJxK contingency table
I×J×K 列联表中的交互作用检验AI 解读 ↓
本页整理公开书目信息,不生成学术质量评分。AI 解读根据题名及所注明的摘要或原文片段生成,未完成全文核验,可能有误;具体假设、证明与结论请核对原文。摘要保留来源文本,未经模型改写。
DOI 首次登记不等于接收或发表日期。部分更正、评论等条目没有独立摘要;缺少可靠材料时保留待补状态。
引用量采用 OpenAlex 单条关联记录,不合计多个记录;关联记录可能包含正式发表版本,不等于此版本单独引用。缺失不视为零,引用指标不代表论文质量。
关注作者按书目中的完整姓名匹配,同名和姓名变体需人工辨别。相关论文先按共享细分方向和具体方法排列,再参考通用方法与发表日期。
AI 根据所提供的论文材料归类:摘要讨论列联表交互检验的有效性,未涉及离散选择行为模型或其他所列细分方向。
反馈分类问题I×J×K 列联表中的交互作用检验AI 解读 ↓
正在核对关联版本…
正在核对对应版本的资源…
Zelen (1971) proposed an asymptotic test for zero three-way interaction in the 2 × 2 × K table. This paper presents two examples for the 2 × 2 × 2 case where Zelen's test directly contradicts the data. We farther show that the correct asymptotic representation of Zelen's exact sampling distribution does, in fact, differ from his result. His test statistic will reject the null hypothesis with probability one asymptotically even under the null hypothesis, except in special instances. The test statistic arising naturally from the asymptotic distribution is the classical goodness-of-fit test obtained by Bartlett (1935) and Norton (1945) and given by Plackett (1974). This result extends to the generalizations of Zelen's result to I × J × K tables as discussed by Gart (1972) and Patdl (1974).