Flexible sequential or sliced designs for computer emulations
报 告 人:: 艾明要
报告地点:: 数学与统计学院415报告厅
报告时间:: 2018年05月18日星期五10:00-11:00

Sequential experiments composed of initial experiments and follow-up experiments are widely adopted for economical computer emulations in complex systems. Many kinds of Latin hypercubes with good space-filling properties have been constructed for designing the initial computer experiments. However, few works based on Latin hypercubes have focused on the design of the follow-up experiments. Although some constructions of nested Latin hypercube designs can be adapted to sequential designs, the size of the follow-up experiments needs to be a multiple of that of the initial experiments. In this paper, a general method for constructing sequential designs of flexible size is proposed, which allows the combined designs to have good one-dimensional space-filling properties. Moreover, the sampling properties and central limit theorems are derived for these designs. Several improvements of these designs are made to achieve better space-filling properties. The related issues for flexible sliced designs are also addressed. Some simulations are carried out to verify the theoretical results.

发 布 人:吴双 发布时间: 2018-06-13
北京大学数学科学学院统计学教研室主任、教授、博士生导师,兼任中国现场统计研究会副秘书长,中国现场统计研究会试验设计分会理事长、高维数据统计分会副理事长、空间统计分会秘书长,是国际统计期刊《Statistica Sinica》、《Journal of Statistical Planning and Inference》、《Statistics and Probability Letters》、《STAT》的Associate Editor,国内数学期刊 《系统科学与数学》的编委。 主要从事试验设计与分析、计算机试验、高维数据分析和应用统计的教学和研究工作,在Ann. Statist.、JASA、Biometrika、Technometrics、Statist. Sinica等国内外期刊发表学术论文六十余篇,主持或参与完成多项国家自然科学基金面上项目、重点项目和国家科技部973课题等。