报告题目一： Inferences for Partially Conditional Quantile Treatment Effect Model

时间：2020年5月14日（周四）下午3:00-3:50

线上平台：腾讯会议

主办单位：数学与统计学院

报告人：唐盛芳 博士

报告人简介：唐盛芳，厦门大学统计学博士，研究方向为理论计量经济学与统计学。

内容简介: In this paper, a new model, termed as the partially conditional quantile treatment effect (PCQTE) model, is proposed to characterize the heterogeneity of treatment effect conditional on some predetermined variable(s). We show that the partially conditional quantile treatment effect is identified under the assumption of selection on observables, which leads to a semiparametric estimation procedure in two steps: first, parametric estimation of the propensity score function and then, nonparametric estimation of conditional quantile treatment effect. Under some regularity conditions, the consistency and asymptotic normality of the proposed semiparametric estimator are derived. More importantly, a specification test is seminally proposed in quantile regression literature, to test whether there exits heterogeneity for PCQTE across sub-populations based on the Cram ́er- von Mises type criterion. The asymptotic properties of the proposed test statistic are investigated, including consistency and asymptotic normality. Finally, the performance of the proposed methods is illustrated through Monte Carlo experiments and an empirical application on estimating the effect of the first-time mother’s smoking during pregnancy on the baby’s birth weight conditional on mother’s age and testing whether the partially conditional quantile treatment effect varies across different mother’s age.

报告题目二：Backward doubly stochastic Volterra integral equations and applications

时间：2020年5月14日（周四）下午3:50-4:40

线上平台：腾讯会议

主办单位：数学与统计学院

报告人：温家强 博士

报告人简介：温家强，山东大学金融数学与金融工程博士，南方科技大学金融数学博士后。

内容简介：Backwardstochasticdifferential equations (BSDEs) have important applications in optimal control problems, financial mathematics, and PDEs. In this talk,Ifirstbriefly review thehistoryof BSDEs. Then, in joint work with Yufeng Shi and Jie Xiong, backward doubly stochastic Volterra integral equations(BDSVIEs) are presented in detail, which is a new type and new development of BSDEs. First, we introduce the notation of Martingale solution and provethe well-posedness of BDSVIEs. Second, a comparison theorem of BDSVIEs is proved. As an application of the comparison theorem, we derive the existence of solutions of BDSVIEs with continuous coefficients. Third, a duality principle between linear forward doubly stochastic Volterra integral equations and BDSVIEs is obtained.Finally, by virtue of the duality principle, a maximum principle of Pontryagin’stype is established for an optimal control problem ofBDSVIEs.

报告题目三：1范数向量值再生核巴拿赫空间中的多任务学习

时间：2020年5月14日（周四）下午4:40-5:30

线上平台：腾讯会议

主办单位：数学与统计学院

报告人：林荣荣 副研究员

报告人简介：林荣荣，中山大学数据科学与计算机学院副研究员，中山大学计算数学专业博士。研究方向为机器学习核函数方法、傅里叶分析、函数逼近论。

内容简介:再生核希尔伯特空间中的机器学习核函数方法具有坚实的数学理论基础，因而广泛应用于统计学习、数据挖掘和散乱数据逼近等。从2009年以来，再生核巴拿赫空间的构造日益受到关注。再生核巴拿赫空间能诱导稀疏范数，可为稀疏表示学习提供理想的背景核函数空间。受稀疏表示学习和多任务学习这两个研究趋势的启发，我们(i)构造了具有表示定理的1范数向量值再生核巴拿赫空间;(ii)证明了表示定理成立与核函数的勒贝格常数之间的关系;(iii) 给出一些核函数勒贝格常数估计；(iv) 对1范数向量值再生核巴拿赫空间中正则化算法给出了误差学习阶估计；(v) 针对机器学习标准数据库(如手写数字和垃圾邮件分类)验证所提出模型能保证稀疏表示和准确率。

“青年学术论坛”是我院新发起的数学统计学科常态化学术交流平台，论坛旨在通过学术报告、学术沙龙、小型研讨会等形式，为我院青年教师和研究生提供向国内外同行学习交流的机会，展现和分享青年教师的研究成果，引导青年教师相互交流切磋，活跃学院学术氛围，提升学院青年教师在数学和统计学科的创新能力。