Functionally Induced Priors for the Analysis of Experiments
Functionally Induced Priors for the Analysis of Experiments
复制标题
DOI：
10.1198/004017006000000372
复制
发表时间：
2007-02
期刊：
影响因子：
2.5
通讯作者：
V. Roshan Joseph;James Dillon Delaney
中科院分区：
文献类型：
--
作者：
V. Roshan Joseph;James Dillon Delaney
这项工作发展的想法，使用功能先验的设计和分析的三个层次和更高层次的实验。开发模型参数的先验分布具有挑战性，因为因子可以是定性的或定量的。我们提出了适当的相关函数和编码方案，使先验分布简单，结果是可解释的。先验结合了众所周知的原则，如效应层次和效应遗传，这有助于几乎自动解决部分设计中的混叠问题。通过对一些真实的实验的分析，说明了新方法的有效性。
This work develops the idea of using functional priors for the design and analysis of three-level and higher-level experiments. Developing a prior distribution for model parameters is challenging, because a factor can be qualitative or quantitative. We propose appropriate correlation functions and coding schemes so that the prior distribution is simple and the results are interpretable. The prior incorporates well-known principles, such as effect hierarchy and effect heredity, which helps resolve the aliasing problems in fractional designs almost automatically. The usefulness of the new approach is illustrated through the analysis of some real experiments.