CONFIDENCE PROCEDURES FOR 2-SAMPLE PROBLEMS
成果类型：
Article
署名作者：
SWITZER, P
署名单位：
Stanford University
刊物名称：
BIOMETRIKA
ISSN/ISSBN：
0006-3444
发表日期：
1976
页码：
1325
关键词：
摘要：
In the usual 2-sample problem one is estimating the constant additive effect of a treatment or the additive constant by which 2 random variables differ. If the treatment-effect may depend on the response level, then a more general approach to 2-sample problems seems appropriate. A treatment-effect function t is defined and distribution-free confidence bounds for the function t are characterized in the case where t has specified parametric forms and in the case where t is not parameterized.