题目：A 2-step kernel method for smoothing spatial point processes
主讲人：虞克明
讲座时间：2026年7月12日(周日) 11：00-12：00
地点：综合楼644会议室
报告人简介：虞克明，英国伦敦布鲁内尔大学统计学与数据科学讲习教授(Chair Professor)、数学学科研究影响中心主任；英国皇家统计学会会士、英国社科基金(ESRC) 评审专家成员、英国自科基金(EPSRC)评审专家成员、欧洲科学基金(ESF) 评审专家成员。目前是《Journal of the Royal Statistical Society-C》的编委，也担任过《Journal of the American Statistical Association, A&CS》、《Journal of the Royal Statistical Society-A》等多家国际SCI、SSCI期刊的编委。他连续入选2021、2022、2023及2025年斯坦福大学/爱思唯尔单年引用影响力排名全球前2%科学家名单。2024年，他被ScholarGPS评为高排名学者。
2026年，他在Research.com评选的全球顶尖数学家排名中位列英国第196位、全球第3054位。虞教授是国际公认的贝叶斯分位数回归先驱，在统计方法论与数据科学领域具有深远影响。先后在《Journal of American Statistical Association》、《Journal of the Royal Statistical Society: Series B》、《Journal of Econometrics》、《Journal of Business & Economic Statistics》、《Bernoulli》等统计学顶级刊物上发表论文150多篇。
讲座摘要：The intensity function is fundamental to spatial point processes, as it describes how point density varies across space and forms the basis for modelling, inference, and simulation. Kernel smoothing is a widely used non-parametric approach for estimating the intensity function (Diggle 1985), (Baddeley et al. 2022), (Macdonald et al. 2025). However, existing kernel-based estimators may exhibit bias near the boundary and substantial variability. This talk proposes an alternative estimator for the intensity function. The asymptotic bias and variance of the proposed estimator are derived for Poisson processes and are complemented by empirical studies for both Poisson processes and log-Gaussian Cox processes. A rule-of-thumb for the band- width selection is suggested. The results demonstrate that the proposed estimator possesses more desirable mean-integral-squared-error, including reduced variance and improved performance near the boundaries of the observation window.
