[第一场] 题　目：Another Look at Design of Blocked Fractional Factorial Split-Plot Experiments 主讲人：郑清水教授(Department of Statistics,University of California, Berkeley) 时　间：99年6月4日(星期五)上午10:00-10:50 (上午10:50-11:10茶会於交大统计所429室举行) 地　点：交大综合一馆427室 Abstract Design of blocked fractional factorial split-plot experiments was considered by McLeod and Brewster (Technometrics, 2004). Split-plot designs arise, e.g., when some treatment factors require larger experimental units than others since their levels are more difficult to change. Blocking is needed when the experimental runs cannot be conducted under homogeneous conditions. In this talk, I will revisit this subject, and discuss the selection and construction of blocked fractional factorial split-plot designs from a different perspective. [第二场] 题　目：Analysis of Current Status Data with Missing Covariates 主讲人：温启仲教授(淡江大学数学系) 时　间：99年6月4日(星期五)上午11:10-12:00 (上午10:50-11:10茶会於交大统计所429室举行) 地　点：交大综合一馆427室 Abstract Statistical inference based on the right-censored data for proportional hazard (PH) model with missing covariates has received considerable attention, but interval-censored or current status data with missing covariates are not yet investigated. Our study is partly motivated by analysis of fracture data from a cross-sectional study, where the occurrence time of fracture was interval-censored and covariate osteoporosis was not reported for all residents. We assume that the data are realized from a PH model. A nonparametric maximum likelihood estimation is proposed to analyze current status data with missing covariates. Simulation and real data analysis exhibit the nice performance of our proposal. -- --

※ 发信站: 批踢踢实业坊(ptt.cc) ◆ From: 140.113.180.45