演講公告 11/9
各位老師、同學：
本次共有兩場演講，因此時間提早至 10：00 開始，茶會：10：50。 歡迎參加。
清大統計所
清華大學、交通大學
統 計 學 研 究 所
專 題 演 講
題 目: Step-stress tests and some exact inferential results
主講人: Professor Narayanaswamy Balakrishnan
Department of Mathematics and Statistics
McMaster University
時 間: 96年11月9日（星期五）10：00 - 10：50
（上午10：50-11：10茶會於統計所821室舉行）
地 點: 清大綜合三館837室
Abstract
In this talk, I will introduce first various models that are used in the
context of step-stress testing. Then, I will describe the cumulative
exposure model in detail and describe the model under the assumption of
exponentiality. I will then discuss the derivation of the MLEs and their
exact conditional distributions, and then present various methods of
inference including exact, asymptotic and bootstrap methods and compare their
performances. I will develop the results for different forms of censored
data, and then present some illustrative examples. Finally, I will point out
some open problems in this direction.
敬請公佈 歡迎參加
清華大學、交通大學
統 計 學 研 究 所
專 題 演 講
題 目: Covariate-adjusted Matrix Visualization using the Generalized
Association Plots Approach
主講人: 吳漢銘博士 (中研院統計所)
時 間: 96年11月9日（星期五）11：10 - 12：00
（上午10：50-11：10茶會於統計所821室舉行）
地 點: 清大綜合三館837室
Abstract
In this study, we extended the framework of matrix visualization (MV) by
incorporating a covariate adjustment process through the estimation of
conditional correlations. The benefit leads directly to the exploration of
conditional association structures among the subjects or variables which is
not available in the conventional MV. For a discrete covariate, the
conditional correlation is estimated by the within and between analysis. This
procedure decomposes a correlation matrix into the within- and
between-component matrices. The contribution of the covariate effects can
then be assessed through the relative structure of the between-component to
the original correlation matrix while the within-components act as residuals.
When a covariate is of continuous nature, the conditional correlation is
equivalent to the partial correlation under the assumption of a joint normal
distribution. In addition, a z score map is proposed to identify variable
pairs with the most significant differences in correlation before and after a
covariate adjustment. Three data sets are employed as examples. The proposed
method has been implemented as a module of the GAP package which is available
at http://gap.stat.sinica.edu.tw/Software/GAP.
敬請公佈 歡迎參加
--
※ 發信站: 批踢踢實業坊(ptt.cc)
◆ From: 59.115.230.171