当前位置:首页  学术交流

学术讲座【Some Semiparametric Model for Length-biased and Right-censored Data】

时间:2013-11-11浏览:615设置

时间:2013年11月13日(周三)10:00-11:00

地点:成功楼603室

主讲:中国科学院/上海财经大学 周勇教授

主办:数学与计算机科学学院

专家简介:周勇教授,博士生导师,中科院数学与系统科学研究院研究员,并受聘于上海财经大学任统计与管理学院院长。1964年8月生,1994年博士毕业于中国科学院应用数学研究所,香港大学统计与精算系博士后,系长江学者特聘教授、国家杰出青年基金获得者、新世纪百千万人才工程国家级人选,国家973项目评委,中国现场统计研究会环境与资源统计分会理事长,国际标准与技术法规分会常务理事,中国统计学会高等教育分会副会长,中国现场统计学会常务理事,概率统计学会常务理事。曾任美国北卡罗莱纳大学(UNC)生物统计系研究副教授,访问过美国普林斯顿大学、香港中文大学、香港城市大学等知名高校。现任多个国内外知名杂志的编委。
    周勇研究员主要从事统计学研究,在数量金融与风险管理、计量经济学、生存分析和生物统计等方面做出了突出贡献。曾主持或作为主要参加者完成过国家级和省部级项目7项,是国家973重大项目骨干成员(排名第二)和863重大项目主要成员。周勇研究员的研究成果曾发表在国际顶级统计杂志《Annals of Statistics》 (SCI 1区),《Journal of the American Statistical Association》(SCI 1区)和顶级计量经济学杂志《Journal of Econometrics》(SSCI 2区)等权威杂志上。论文被SCI引用212次,SCI他引190次。

报告摘要:Length-biased sampling data are often encountered in the studies of economics, industrial reliability, etiology applications epidemiology, genetics and cancer screening. The complication of this type of data is due to the fact that the observed lifetime suffers from left truncation and right censoring, where the left truncation variable has a uniform distribution. In this talk, we intend to study the accelerated failure time model (AFT) with length-biased sampling data by using the composite partial likelihood technique (Huang and Qin, 2012). The proposed method has a very simple form. To ease the calculations for estimates, we use a kernel smoothed estimation method (Heller, 2007). Large sample results and a re-sampling method for the variance estimation are discussed. A simulation study is conducted to compare the performance of the proposed method with other existing methods. A real data set is used for illustration. Furthermore, we considers the monotonic transformation model with unspecified transformation function and unknown error function, and gives its  monotone rank estimation  with length-biased and right-censored data. The estimator is shown to be  -consistent and asymptotically normal. Numerical simulation studies reveal good finite sample performance. The variance could be estimated by a resampling method via perturbing the U-statistics objective function repeatedly, which avoids the choice of smoothing parameters by using numerical derivatives.

返回原图
/