时间:2016年11月8日(星期二)下午15:00—16:00
地点:史带楼502室
主持人:张新生 教授 复旦大学管理学院统计学系
主题:Individual Factor Word Length Pattern For Fractional Factorial Designs
主讲人:Professor William Li 复旦大学特聘讲座教授
Xian Dong Eric Jing Professor
Carlson School of Management,University of Minnesota
摘要: Literature on constructing efficient experimental designs has been plentiful, but how best to incorporate prior information when assigning factors to the columns on a nonregular design has received no attention. Following Li et al. (2015), we propose a new criterion called the individual generalized word length pattern (iGWLP) for ranking columns of a nonregular design. A theorem is given to relate the iGWLP to the expected bias caused by model mis-specifications. We also show that the proposed criterion may lead to designs having better projection properties in the factors considered most likely to be important. Taking examples from the literature of recommended orthogonal arrays, we illustrate how iGWLP helps identify important differences in the aliasing that is likely otherwise missed. Given the complexity of characterizing partial aliasing for nonregular designs, iGWLP will help practitioners make more informed assignment of factors to columns when utilizing nonregular fractions. Furthermore, we demonstrate that iGWLP provides a useful supplement to clear interaction graphs.
统计学系
2016-11-7
活动讲座
新闻动态
微信头条
招生咨询
媒体视角
瞰见云课堂