Computer Experiments With Qualitative and Quantitative Factors
主 题: Computer Experiments With Qualitative and Quantitative Factors
报告人: 钱智光 博士 (University of Wisconsin-Madison)
时 间: 2009-07-09 上午10:30 - 11:30
地 点: 理科一号楼 1418
Computer experiments, i.e., complex mathematical models implemented in large computer codes, are getting more and more traction in engineering, marketing and sciences. In this talk, I will give an overview of statistical challenges in the field of computer experiments and then report some advances in the design and modeling of computer experiments with qualitative and quantitative factors. I will present a general approach to constructing a new type of design, called sliced space-filling design, to accommodate both qualitative and quantitative factors. It starts with constructing a Latin hypercube design based on a special orthogonal array for the quantitative factors and then partitions the design into groups corresponding to different level combinations of the qualitative factors. The points in each group achieve good space-filling properties. Also will be discussed is a novel framework for building statistical interpolators with both qualitative and quantitative factors.
