1st 统 计 学 研 究 所 专 题 演 讲 题 目: Kullback-Leibler Information for Model Selection and Comparison in Logistic Linear Models 主讲人: Dr. Guoqi Qian, Department of Mathematics and Statistics, University of Melbourne, Australia 时 间: 96年12月7日（星期五）10：00 - 10：50 （上午10：50-11：10茶会於统计所821室举行） 地 点: 清大综合三馆837室 Abstract We consider two issues involved in model selection. 1. Model selection starts with a proposed model class, and it is often unrealisticlly assumed that the true model generating the data belongs to this model class. Then what would happen to model selection if the true model is not in the proposed model class? In other words, how to quantify the model selection bias in the situation of model class mis-specification? 2. Model selection often ends up with a selected optimum model minimizing or maximizing a numeric selection criterion function. But it does not or is not able to provide a measure of variability or uncertainty involved in model selection. Such a measure, if available, would be very useful in determining models which are indistinguishable from the optimum model. We have developed new estimators of Kullback-Leibler information to address these two issues. Our work will be presented in the context of logistic regression model selection but can be extended to other model selection problems. 敬请公布 欢迎参加 2nd 清华大学、交通大学 统 计 学 研 究 所 专 题 演 讲 题 目: Confidence intervals and confidence bands for the difference of two median survival times 主讲人: 张玉媚教授 (东海大学统计学系) 时 间: 96年12月7日（星期五）11：10 - 12：00 （上午10：50-11：10茶会於统计所821室举行） 地 点: 清大综合三馆837室 Abstract We are concerned with the estimation of the discrepancy between two treatments when right-censored survival data are accompanied with covariates. Conditional confidence intervals given the available covariates and confidence bands over a certain region of covariates are constructed for the difference between two median survival times under the unstratified and stratified Cox proportional hazards models, respectively. The proposed confidence intervals and confidence bands provide the information about the difference in survivorship for patients with common covariates but in different treatments. The results of the simulation study investigation of the coverage probability and expected length of the confidence intervals and bands suggest the one designed for the stratified Cox model when data fit reasonably with the model. When the stratified Cox model is not feasible, however, the one designed for the unstratified Cox model is recommended. The use of the confidence intervals and confidence bands is finally illustrated with a HIV+ data set. 敬请公布 欢迎参加 --

※ 发信站: 批踢踢实业坊(ptt.cc) ◆ From: 140.113.114.185 ※ 编辑: mangogogo 来自: 140.113.114.185 (12/07 09:29)