报告时间：2026年6月11日（周四）下午 15:30-16:30
报告地点：苏州大学天赐庄校区精正楼306
报告人：尤翀 副教授，复旦大学与上海数学与交叉学科研究院
报告摘要：
This paper proposes a unified inferential framework for high-dimensional mixed-type multi-response regression, accommodating continuous, binary, count, and survival outcomes. By extending classical multivariate test statistics—such as Wilks' lambda, the Hotelling–Lawley trace, and Pillai's trace—we develop a residual-projection-based testing procedure that leverages cross-response dependencies for joint inference. The method is theoretically justified with asymptotic validity and demonstrates in simulations accurate Type I error control and competitive power.
报告人简介：
尤翀，复旦大学与上海数学与交叉学科研究院双聘副教授，主要研究兴趣包括变分贝叶斯、动态个体化治疗、变量选择、疾病预测以及传染病建模等领域。研究重点是运用现代统计方法，解决高维数据分析、生命健康预测与精准医疗中的重要问题。
邀请人：徐礼柏