实事求是 经世致用

宁波大学logo

甬江数学讲坛(2019年第14讲)

2019-06-04

报告题目:On  Performances of Some Algorithms for Multivariate Data

报告人:  Jin Wang (王晋)  副教授  

Northern  Arizona University, USA (美国北亚利桑那大学)  


报告时间:
20190606  (星期四)  9:30-10:30


报告地点:龙赛理科楼北楼
116

报告摘要:In  modern statistics, almost all statistical methods are implemented through  algorithms. Thus performance of a statistical method is directly affected by the  algorithm for the method. Here we study two well-known algorithms for  multivariate data. It is found that the performances of both algorithms decline  as dimension increases. The effect of data shape on the performances of the  algorithms is also studied. Our finding is that the performances of both  algorithms decrease as kurtosis increases. Some adjustments for those algorithms  will be discussed as well, along with some new descriptive measures for  multivariate data.