Construction of sliced maximin-orthogonal Latin hypercube designs
Construction of sliced maximin-orthogonal Latin hypercube designs
复制标题
DOI：
10.5705/ss.2013.352
复制
发表时间：
2016-04
影响因子：
1.4
通讯作者：
Jinyu Yang;Hao Chen;D. Lin;Min-Qian Liu
中科院分区：
文献类型：
--
作者：
Jinyu Yang;Hao Chen;D. Lin;Min-Qian Liu
切片拉丁超立方体设计是一种特殊的拉丁超立方体设计，它可以被划分成更小的拉丁超立方体设计的切片。这种类型的设计对于具有定性和定量因素的计算机实验、多个实验、数据池和交叉验证非常有用。拉丁超立方体设计的可拓性和均匀性是其重要性质。本文利用正交设计、Goethals-Seidel阵列和Kharaghani阵列构造了切片极大极小正交拉丁超立方体设计。所得设计具有二阶正交性和良好的均匀性。
A sliced Latin hypercube design is a special Latin hypercube design that can be divided into slices of smaller Latin hypercube designs. This type of designs is useful for computer experiments with qualitative and quantitative factors, multiple experiments, data pooling, and cross-validation. Orthogonality and uniformity are important properties for Latin hypercube designs. In this paper, sliced maximin-orthogonal Latin hypercube designs are constructed using orthogonal designs, Goethals-Seidel arrays, and Kharaghani arrays. The resulting designs have both second-order orthogonality and good uniformity.