报告人：喻达磊（云南财经大学）

时间：2022年10月14日 10:30-

腾讯会议ID：958 845 258

摘要：This paper considers the problem of post-averaging inference for model averaging estimators in generalized linear model (GLM). We establish the corresponding asymptotic distributions of frequentist model averaging estimators for GLMs. Depending on the configuration of the penalty term in the weight choice criterion, the asymptotic distribution of the model averaging estimator is not standard or normal. In addition, we propose a feasible simulation-based confidence interval and its asymptotic properties are also investigated rigorously. Monte Carlo simulations further verify the usefulness of our theoretical results, and the proposed methods are also employed to analyze the stock car racing dataset.

简介：喻达磊，博士(香港城市大学)，云南财经大学教授，博士生导师。研究领域为随机效应模型、混合模型以及空间计量模型的模型选择、模型平均和估计理论等。已在包括JRSS-B，JASA，JBES和中国科学：数学在内的国内外统计学期刊上发表论文十余篇。主持国家自然科学基金项目三项，入选了云南省中青年学术和技术带头人后备人才和云南省“万人计划”的“青年拔尖人才”专项。

邀请人：夏小超

欢迎广大师生积极参与！