Matthews J N S
Department of Statistics, University of Newcastle, Newcastle upon Tyne, UK.
Dev Biol (Basel). 2002;107:87-94.
Limiting dilution assays are a useful device for extracting quantitative information when the underlying analytical tool can only provide information on presence or absence. Analysis by maximum likelihood is advocated and some alternative methods are discussed. Several interval estimators, including one based directly on the log-likelihood are presented. Methods for determining designs which permit efficient estimation are considered and some of their weaknesses are mentioned. Opportunities for further methodological development are outlined.
当基础分析工具只能提供存在或不存在的信息时，有限稀释分析是提取定量信息的一种有用方法。提倡采用最大似然分析，并讨论了一些替代方法。提出了几种区间估计方法，包括一种直接基于对数似然的方法。考虑了确定允许有效估计的设计的方法，并提及了它们的一些弱点。概述了进一步方法学发展的机会。
