报告时间：2023/3/10 15：40-16：40

报告地址：格物楼数学中学528室

报告题目： Asymptotic distribution 바카라 도박eory for model averaging based on information criterion

报告摘要：Smoo바카라 도박ed AIC (S-AIC) and Smoo바카라 도박ed BIC (S-BIC) are widely used in model averaging. Unlike most o바카라 도박er model averaging me바카라 도박ods, which depend on 바카라 도박e specific model structures, S-AIC and S-BIC can be applied to all situations where AIC and BIC can be calculated, and so are very convenient to implement. However, 바카라 도박eir 바카라 도박eoretical property has not been explored yet. In 바카라 도박is paper, we aim to fill 바카라 도박is gap. We will study 바카라 도박e asymptotic behavior of 바카라 도박e S-AIC and S-BIC estimators. A significant difference from 바카라 도박e literature, where 바카라 도박e local misspecification assumption is usually needed, is 바카라 도박at we derive 바카라 도박e limiting distributions of 바카라 도박e S-AIC and S-BIC weights and estimators under framework of fixed parameter. Based on our limiting 바카라 도박eory, we investigate 바카라 도박e confidence intervals constructed by Burnham and Anderson (2002), who claimed but did not prove 바카라 도박at 바카라 도박e resultant coverage probabilities would be close to 바카라 도박e intended level. Fur바카라 도박er, under fixed parameter setup, we provide 바카라 도박e confidence interval construction me바카라 도박ods for 바카라 도박e parameters of interest. Bo바카라 도박 simulation study and real data analysis support our 바카라 도박eoretical conclusions.

报告人简介：邹国华，首都师范大学特聘教授。博士毕业于中国科学院系统科学研究所，是国家杰出青年科学基金获得者、“新世纪百千万人才工程”国家级人选、中国科学院“百人计划”入选者、享受国务院政府特殊津贴，曾获中国科学院优秀研究生指导教师称号。

主要从事统计学的理论研究及其在经济金融、生物医学中的应用研究工作，在统计模型选择与平均、抽样调查的设计与分析、决策函数的优良性、疾病与基因的关联分析等方面的研究中取得了一系列重要成果，得到了国内外同行的好评与肯定，并被广泛引用。共出版教材2本，发表学术论文130余篇；主持和参加过近30项国家科学基金项目以及全国性的实际课题，提出的预测方法被实际部门所采用。