摘要
沈清文［１］和李开灿［２］分别提出了ＩｘＪｘＫ列联表中辅助交互作用的强可压缩性．然而，该概念对有序背景变量是不适合的．这是因为在实际中人们只会连续地合并有序背景的水平，而不会把不相邻的若干个水平合并成一组．为此，本文提出了辅助交互作用对有序背景变量的有序可压缩性，并得到了其充分必要条件。此条件在背景变量连续水平的可能合并中起着重要的指导作用．
Shen(1991) and Li(1998) presented the strong collapsibility of additive interaction in I x J x K tables.. However, tile concept is not suitable for an ordinal background variable since it is meaningless to pool disjoint levels in this case. In this paper, we introduce the consecutive collapsibility of additive interactions over an ordinal background and present necessary and sufficient conditions for consecutive collapsibility of additive interactions. The conditions are important for guiding the possible merging of consecutive levels of tile background.
出处
《应用概率统计》
CSCD
北大核心
2001年第1期39-43,共5页
Chinese Journal of Applied Probability and Statistics
基金
国家自然科学基金资助项目!(项目编号:10001008)
