Sequentially Refined Latin Hypercube Designs: Reusing Every Point
Sequentially Refined Latin Hypercube Designs: Reusing Every Point
复制标题
DOI：
10.1080/01621459.2014.993078
复制
发表时间：
2015-10
影响因子：
3.7
通讯作者：
Jin Xu;Jiajie Chen;Peter Z. G. Qian
中科院分区：
文献类型：
--
作者：
Jin Xu;Jiajie Chen;Peter Z. G. Qian
使用迭代扩大的拉丁超立方体设计运行计算机实验最近在实践中得到普及。这种方法使用拉丁超立方体设计的计算机代码进行初始实验，然后运行精心选择的额外运行的后续实验，使得两个实验的组合设计集形成更大的拉丁超立方体设计。这个增广过程可以重复多个阶段，其中在每个阶段中，增广设计集保证是拉丁超立方体设计。我们提供了一个理论框架，把这种方法的坚实基础。数值例子证实了理论结果。本文的补充材料可在网上查阅。
The use of iteratively enlarged Latin hypercube designs for running computer experiments has recently gained popularity in practice. This approach conducts an initial experiment with a computer code using a Latin hypercube design and then runs a follow-up experiment with additional runs elaborately chosen so that the combined design set for the two experiments forms a larger Latin hypercube design. This augmenting process can be repeated multiple stages, where in each stage the augmented design set is guaranteed to be a Latin hypercube design. We provide a theoretical framework to put this approach on a firm footing. Numerical examples are given to corroborate the derived theoretical results. Supplementary materials for this article are available online.