THE ESTIMATION OF EXTREME QUANTILES IN LOGIT BIOASSAY
成果类型：
Article
署名作者：
MCLEISH, D; TOSH, D
署名单位：
Acadia University
刊物名称：
BIOMETRIKA
ISSN/ISSBN：
0006-3444
DOI：
10.1093/biomet/70.3.625
发表日期：
1983
页码：
625632
关键词：
摘要：
Suppose a low dose of a drug is administered and, if no response is observed, increase this dose by a fixed amount which is administered to a 2nd experimental subject. This procedure continues until, at some dose level, a response is finally observed. This whole sequence may be repeated. The features of an experimental design of this kind for estimating the dose at which the expected response is small were investigated. This design is highly efficient when the the number of responses, e.g., deaths, is considered the major portion of the cost of the study, and various alternatives having some advantage over maximum likelihood estimation, are suggested.
来源URL：