• 统计研究中心
当前位置: 首页 > 系列讲座 > 正文

拉夫堡大学刘鹏博士:Functional Linear Quantile Regression on a Two-dimensional Domain二维域上的泛函线性分位数回归



主 题Functional Linear Quantile Regression on a Two-dimensional Domain二维域上的泛函线性分位数回归

主讲人拉夫堡大学刘鹏博士

主持人统计学院吴量副教授

时间:202496日(周五)上午1030-1130

举办地点:柳林校区弘远楼408会议室

主办单位:统计研究中心和统计学院 科研处

主讲人简介:

Dr. Peng Liu is a Senior Lecturer in Statistics at Department of Mathematical Sciences, Loughborough University (2024 - Now). Previously he was a Lectuer in Statistics at School of Mathematics, Statistics and Actuarial Science, University of Kent (03/2019 - 08/2024), a Postdoctoral Fellow at Department of Mathematical and Statistical Sciences, University of Alberta (11/2017 - 12/2018), a Senior Postdoctoral Fellow at Department of Biostatistics, University of Washington (08/2015 - 07/2017) and Fred Hutchinson Cancer Research Centre (08/2015 - 10/2017), a Research Assistant at Department of Mathematics, Hong Kong Baptist University (11/2013 - 02/2015). He has a PhD in Biostatistics from Academy of Mathematics and Systems Science, Chinese Academy of Science (2015) and BSc degree from Central China Normal University (2010).

His research interests include biostatistics, functional and neuroimaging data analysis, and machine learning. He has received the Institute of Mathematical Statistics (IMS) New Researchers Conference Travel Award (2018). His research has been supported by UK Research and Innovation (UKRI), Innovate UK, the Academy of Medical Sciences (UK) and Kent Institute of Cyber Security for Society. He is a Leading Guest Editor of the Special Issue on Explainable AI for Industrial Information Integration in Industrial IoT (IIoT) of the journal 'Internet of Things' (02/2024-Now). He is the Editor in Chief for the journal International Journal of Organizational and Collective Intelligence (IJOCI) (01/2024-Now).

刘鹏博士现为拉夫堡大学数学科学系的统计学高级讲师(2024年至今)。此前,他曾任肯特大学数学、统计与精算科学学院的统计学讲师(20193月至20248月),阿尔伯塔大学数学与统计科学系的博士后研究员(201711月至201812月),华盛顿大学生物统计学系和弗雷德·哈钦森癌症研究中心的高级博士后研究员(20158月至20177月),以及香港浸会大学数学系的研究助理(201311月至20152月)。他于2015年在中国科学院数学与系统科学研究院获得生物统计学博士学位,并于2010年在华中师范大学获得理学学士学位。

他的研究兴趣包括生物统计学、神经影像数据分析以及机器学习。他曾获得国际数理统计学会(IMSNew Researchers Conference Travel Award2018年)。他的研究得到了英国国家科研与创新署UKRI)、英国创新署、英国医学科学院以及肯特网络安全与社会研究所的支持。他现任期刊《物联网》工业物联网(IIoT)领域的“工业信息集成中的可解释人工智能”特刊的客座主编(20242月至今),同时也是期刊《Journal of Organizational and Collective Intelligence》(IJOCI)的主编(20241月至今)。

内容简介

This article considers the functional linear quantile regression which models the conditional quantile of a scalar response given a functional predictor over a two-dimensional domain. We propose an estimator for the slope function by minimising the penalised empirical check loss function. Under the framework of reproducing kernel Hilbert space, the minimax rate of convergence for the regularised estimator is established. Using the theory of interpolation spaces on a two- or multi-dimensional domain, we develop a novel result on simultaneous diagonalisation of the reproducing and covariance kernels, revealing the interaction of the two kernels in determining the optimal convergence rate of the estimator. Sufficient conditions are provided to show that our analysis applies to many situations, for example, when the covariance kernel is from the Mat\'ern class, and the slope function belongs to a Sobolev space. We implement the interior point method to compute the regularised estimator and illustrate the proposed method by applying it to the hippocampus surface data in the ADNI study.

这篇文章将探讨函数线性分位数回归,该回归模型用于描述在给定二维域上的函数型预测变量条件下的标量响应的条件分位数。主讲人通过最小化带有惩罚项的经验检查损失函数,提出斜率函数的估计方法。在再生核Hilbert空间框架下,确立正则化估计的最小最大收敛速度。利用二维或多维域上的插值空间理论,主讲人开发一项关于再生核与协方差核的同时对角化的新结果,揭示这两个核在确定估计器的最优收敛速度方面的相互作用。主讲人提供充分条件,证明他们的分析适用于多种情况,例如当协方差核属于Matérn类且斜率函数属于Sobolev空间时。主讲人实施内点法来计算正则化估计,并通过在ADNI研究中的海马表面数据进行应用,展示了所提出方法的有效性。


上一条:麦吉尔大学张启煌助理教授:具有响应测量误差的零膨胀泊松模型

下一条:伦敦政治经济学院Qiwei Yao教授:Probabilistic Forecasting for Daily Electricity Loads and Quantiles for Curve-to-Curve Regression