当前位置: 首页  >  学术报告  >  正文

学术报告

2021年数学学院“吉大学子全球胜任力提升计划”研究生系列短课程(16)

发布时间:2021-07-08 15:01:44   |   点击数量:

报告题目:Some statistical methods for biomedical research

报 告 人:Jin Zhezhen,Columbia University

报告地点:Zoom 会议 ID:927 025 8459

校内联系人:王培洁 wangpeijie@jlu.edu.cn


Abstract: This lecture will be based on my own experience in biomedical research. I will review some commonly used statistical methods in biomedical studies and present related recent research development and topics. Topics include measures of agreement, item selection methods, time to event analysis for transplantation, semiparametric modeling approaches and current challenges and issues for the analysis of big data.


授课日期

Date of Lecture

课程名称(讲座题目)

Name (Title) of Lecture

授课时间

Duration (Beijing Time)

参与人数

Number of Participants

July 12, 2021

Measures of agreement

9:00-10:00

60

July 13, 2021

Item selection

9:00-10:00

60

July 15, 2021

Time to event analysis

9:00-10:00

60

July 16, 2021

Semiparametric   regression models for censored data

9:00-10:00

60

July 17, 2021

General semiparametric   regression models

9:00-10:00

60

July 18, 2021

Big data analysis

9:00-10:00

60


Lecture 1: Measure of agreement

In practice, it is important to examine agreement among measures obtained by different sources or methods. Several commonly used statistical methods for agreement will be reviewed, Bland Altman method, coefficient of variation, mean squared deviation, total deviation index, concordance probability, correlation coefficient (Pearson and Spearman), intraclass correlation coefficient.  

Lecture 2: Item selection

Item selection arises in studies with questionnaire. I will discuss non-parametric approach for the selection of items in a scale for screening, with the score defined as the sum of item response indicators. Without specifying parametric models for binary classification probabilities, the item selection method evaluates the change in classification accuracy due to adding or deleting one item for a scale with k items.

Lecture 3: Time to event analysis

With the transplant examples, basic concepts for time to event analysis and commonly used statistical methods will be discussed.

Lecture 4: Semiparametric regression models for censored data

Semiparametric regression models for censored data will be reviewed along with challenges and available methods.

Lecture 5: General semiparametric regression models

General semiparametric models will be discussed along with inference procedures.

Lecture 6: Big data analysis

Types of commonly encountered big data, and available statistical approaches will be discussed.


报告人简介:Jin Zhezhen,美国哥伦比亚大学梅尔曼公共卫生学院生物统计学系终身教授,长期以来从事生存分析,重采样方法,纵向数据分析、非参数和半参数建模等方向的研究,同时在心脏病学,神经病学,癌症, 阿尔茨海默氏病和流行病学领域也进行了合作研究。1989年毕业于南开大学,1992年获得南开大学硕士学位,1994年美国南加州大学应用数学硕士,1998年哥伦比亚大学博士学位,1998-2000年哈佛大学公共卫生学院博士后,哈佛大学艾滋病研究生物统计学中心高级统计员,2004年香港科技大学数学系访问教授。担任Journal of American Statistical Association—Application and Case Studies, Statistica Sinica, Lifetime Data Analysis, Kidney International, Journal of Statistical Theory and Practice, Communications for Statistical Applications and Methods等杂志的副主编。共计发表SCI论文200多篇。