当前位置:首页 > 中文 > 系列讲座

英国约克大学 Prof. Wenyang Zhang:A New Approach of Multilevel Modelling for Clustered Survival Data

发布时间:2017-11-30

光华讲坛——社会名流与企业家论坛第4772期

 

主 题:A New Approach of Multilevel Modelling for Clustered Survival Data

主讲人:Prof.Wenyang Zhang

主持人:郭斌博士

时 间:2017年12月01日(星期五)上午10:30-11:30

地 点:弘远楼会议室402B

主办单位:统计研究中心 统计学院  科研处

 

主讲人简介:

Prof.Wenyang Zhang,本科毕业于四川大学数学系,后师从统计大师范剑青教授。主要从事非参数统计、时间序列分析、生存分析等方向的研究。曾先后在英国伦敦政治经济学院、英国Kent大学、英国Bath大学、英国York大学任教,现为英国York大学统计学首席教授。他曾是英国皇家统计学会科研委员会委员(历史上仅有三位华人担任该委员会委员),目前是统计学四大国际顶尖期刊之一Journal of the American Statistical Association的副主编。在《Journal of the American Statistical Association》、《The Annals of Statistics》、《Journal of the Royal Statistical Society, Series B》等国际一流期刊发表论文数十篇。

其个人主页为:https://www.york.ac.uk/maths/staff/wenyang-zhang/

 

主要内容:

Inmultilevel modelling for clustered survival data, to account for the difference among different clusters, a commonly used approach is to introduce cluster effects, either random or fixed, in the modelling.  The modelling with random effects may lead to difficulties in the implementation of the estimation procedure for the unknown parameters of interest, because numerical computation for multiple integral may become unavoidable when the cluster effects are not scalars.  On the other hand, if fixed effects are used, there would be a danger of having estimators with big variances, because there are too many nuisance parameters involved in the model used.  In this talk, I will show a new approach of multilevel modelling for clustered survival data.  The proposed modelling does not have the potential computational problem which the modelling with random effects does, it also involves far less unknown parameters than the modelling with fixed effects.  Some asymptotic properties of the proposed modelling and intensive simulation study results will be presented to demonstrate the advantage of the proposed method.  Finally, I will apply the proposed method to analyse a data set about the second-birth interval in Bangladesh.  The findings are quite interesting.