会议时间：2022/10/11 18:30-21:30 (GMT+08:00)中国标准时间-北京

点击链接入会，或添加至会议列表：https://meeting.tencent.com/dm/6gIFsfazIXeZ

腾讯会议：791-762-620

会议密码：202210

Title:On the use of degrees of freedom in linear model with repeated measurement errors

Abstract:Degrees of freedom (DF) is a fundamental concept in statistical modeling and plays important roles in quantitative description of model complexity, unbiased estimation of the prediction error, and model assessment and selection. However, the existence of measurement errors changes the value of DF and makes the calculation not so intuitive. In linear measurement error models with replicates, one can construct unbiased estimating equations based on the independence of two replicates, and then obtain a consistent estimator. In this context, we propose unbiased estimators of DF and apply it to information criterion construction for model selection and model averaging, respectively. We establish asymptotic properties and conduct simulations for evaluating their finite sample performances.

个人简介：柏杨，上海财经大学统计与管理学院教授（长聘），博士生导师；上海财经大学数据科学与统计研究院校聘研究员，副院长；中国现场统计研究会资源与环境统计分会副理事长。柏杨博士一直从事纵向数据分析等统计理论和方法的科学研究工作，在复杂数据分析与建模方面开展了许多有意义的研究工作。目前在Statistica Sinica、Scandinavian Journal of Statistics、Journal of Multivariate Analysis等国际知名统计学期刊上发表论文20余篇，google 学术他引150余次。已经主持完成国家自然科学基金面上项目和青年项目各一项，第三完成人获得教育部2014年度高等学校科学研究优秀成果奖二等奖1项。