Southwest Jiaotong University School of Mathematics

统计系

首页  >  学术科研  >  统计系  >  正文

西南交大2021年统计学系列学术报告(成渝经济圈统计学术交流)

来源:统计学   作者:统计学     日期:2021-11-18 21:12:37   点击数:  

报告信息--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

开始时间:2021年11月20日,上午9:00时整开始

报告形式:线上腾讯会议

会议时间: 2021/11/20 09:00-13:00 (GMT+08:00)

点击专属链接入会,或添加至会议列表:

https://meeting.tencent.com/dw/bwykgSITDbn4

会议 ID:281 755 831

会议密码:211120




报告I(9:10-9:55,报告环节约40分钟,提问环节5分钟)

题目: Testing the martingale difference hypothesis in high dimension

摘要:In this paper, we consider testing the martingale difference hypothesis for high-dimensional time series. Our test is built on the sum of squares of the element-wise max-norm of the proposed matrixvalued nonlinear dependence measure at different lags. To conduct the inference, we approximate the null distribution of our test statistic by Gaussian approximation and provide a simulation-based approach to generate critical values. The asymptotic behavior of the test statistic under the alternative is also studied. Our approach is nonparametric as the null hypothesis only assumes the time series concerned is martingale difference without specifying any parametric forms of its conditional moments. As an advantage of Gaussian approximation, our test is robust to the panel dependence of unknown magnitude. To the best of our knowledge, this is the first valid test for the martingale difference hypothesis that not only allows for large dimension but also captures nonlinear serial dependence. The practical usefulness of our test is illustrated via simulation and a real data analysis. The test is implemented in a user-friendly R-function.

报告人简介:

常晋源教授于2013年7月在北京大学光华管理学院取得经济学博士学位,2013年9月至2017年2月在澳大利亚墨尔本大学数学与统计学院任研究员,2017年3月开始全职在西南财经大学统计学院工作。现为西南财经大学数据科学与商业智能联合实验室执行主任、教授、博士生导师,国家杰出青年科学基金获得者,国家青年“QR计划”特聘专家和教育部青年长江学者,四川省特聘专家、四川省统计专家咨询委员会。主要从事“超高维数据分析”和“高频金融数据分析”两个领域的研究。先后以第一作者在《Annals of Statistics》《Biometrika》《Biometrics》和《Journal of Econometrics》等统计学与计量经济学国际顶级学术期刊发表论文十余篇。现目前正担任统计学国际顶级学术期刊Journal of the Royal Statistical Society Series B、统计学国际知名学术期刊Statistica Sinica以及计量经济学国际著名学术期刊Journal of Business & Economic Statistics的Associate Editor。



报告II(09:55-10:40,报告环节约40分钟,提问环节约5分钟)

题目: Robust and Joint Feature Screening in High Dimension

摘要:In this talk, I will give an introduction to ultrahigh-dimensional robust and joint feature screening. Specifically, I first review the seminal work of Fan and Lv (2008).  Then, I will present three concrete robust and joint feature screening approaches including: (a) the conditional quantile correlation-based sure independence screening (CQC-SIS) under varying coefficient models, (b) the copula partial correlation-based sure independence screening (CPC-SIS), and (c) a robust partial correlation-based sure independence (RPC-SIS), according to our recent work. For each approach, I will talk about the motivation, methodology, and main theoretical results. Some empirical results would be shown if time permits.

报告人简介:

夏小超,博士,重庆大学数学与统计学院硕士生导师、副教授,研究兴趣:高维特征筛选、模型平均预测、非参半参模型。正在主持1项国家自然科学青年基金项目和1项中央高校基金项目,主持并完成1项湖北省自然科学基金项目和1项中央高校基金项目,在国际知名刊物Journal of Econometrics, Biometrics、Statistica Sinica、Scandinavian Journal of Statistics、Computational Statistics & Data Analysis等多个SCI杂志上发表和录用近二十篇论文,任美国数学评论(Mathematical Reviews)评论员,为多个SCI杂志匿名审稿人。



报告III(10:40-11:25,报告环节约40分钟,提问环节约5分钟)

题目: 高维两样本均值检验

摘要:Permutation tests are widely used in practice. However, these tests either need restrictive assumptions for the validity, or are not applicable to high-dimensional data. This paper considers permutation tests for high-dimensional mean comparison, where in order to get round those restrictions, the test statistics are calculated based on pseudo samples that are generated through a “binning" procedure. The corresponding permutation tests are proved to be asymptotically consistent. We also consider a related problem for signal identication and establish the asymptotic properties. Simulation studies demonstrate favorable performance of our methods in comparison to existing tests. Finally, the pro-posed method is applied to a genome-wide association study (GWAS) for seven complex human diseases to identify possible single nucleotide polymorphisms(SNPs) associated with the diseases.  

报告人简介:

孔婀芳教授博士毕业于新加坡国立大学,先后在荷兰埃因霍芬大学做博士后,英国肯特大学讲师,现就职于电子科技大学数学科学学院,博士生导师,国家青年“QR计划”特聘专家。主要研究方向是非参数半参数模型。近年来主要从事稳健估计、大数据降维、计量经济回归等理论与应用问题的研究,做出了有特色的创新性研究成果,受到国内外同行专家的广泛关注和重视。孔教授在国际顶级学术期刊《Journal of the American Statistical Association》,《Annals of Statistics》,《Biometrika》以及《Econometric Theory》等发表论文十余篇。