报告人: 范协铨 副教授

报告人单位: 天津大学

报告时间:4月12日上午9点

摘 要: Abstract: Self-normalized Cramer type large deviations for independent random variables has been well studied in recent year. One of the most interesting work is due to Jing, Shao and Wang (2003, Ann. Probab.). They proved that self-normalized Cramer type large deviation results hold only under a finite $(2+\rho)$th moment, $0<\rho\leq 1.$ However, we are not aware of any such results in the literature regarding martingales. In this talk, we present some extensions for the results of Jing, Shao and Wang to the martingale case. An application to Student's t-statistic is also discussed. (This talk is based on join work with Ion Grama, Quansheng Liu and Qi-Man Shao)

报告人简介:

范协铨，天津大学，应用数学中心副教授。2009－2013年间，国家公派博士生, 大西洋布列塔尼数学实验室, 南布列塔尼大学, 法国。2013.10-2015.9   博士后, 正规组, 法国国家信息与自动化研究所。2015年9月至今，天津大学副教授。研究兴趣包括大偏差 (Cramér型大偏差; 中偏差) ，重稳定随机过程及其应用，集中不等式 (Berry-Esseen界; 极限理论; 指数不等式) ，鞅; 独立随机变量之和; 经验过程 。其部分成果发表在 STOCHASTIC PROCESSES AND THEIR APPLICATIONS，ELECTRONIC JOURNAL OF PROBABILITY，SCIENCE CHINA-MATHEMATICS 等国际著名学术杂志上。