北航经管学院“经济统计论坛”系列讲座
（2026年第2期，总第39期）
讲座题目：Safety Belt Estimation
讲座时间：2026年6月11日（周四），10:00-11:00
讲座地点：新主楼 A949
主讲人：Dietrich von Rosen 教授
邀请人：康雁飞 教授
讲座嘉宾
Dietrich von Rosen is professor in Mathematical Statistics at Linköping University at the Department of Mathematics, and professor in Statistics at the Swedish University of Agricultural Sciences at the Department of Energy and Technology. The year 1998 von Rosen got a full professorship in Statistics at the Swedish University of Agricultural Sciences (SLU) and 2009-2024 he was also working as adjoint professor in Mathematical Statistics at Linköping University. In year 2024 he formally retired from SLU but is still connected with MAI at Linköping University. Moreover, von Rosen has been working at the medical university in Stockholm, the Karolinska Institute, for more than six years. von Rosen’s research profile comprises linear models, multivariate analysis, bilinear models, high-dimensional analysis and matrix algebra. More than 130 peer reviewed articles have been written and 19 Ph.D. students have under supervision by von Rosen successfully defended their thesis. von Rosen has two published books entitled “Advanced Multivariate Statistics with Matrices” and “Bilinear Regression Analysis, an Introduction”. He is Editor- in- Chief of Journal of Multivariate Analysis. von Rosen has organized several international conferences and workshops. In 2014 von Rosen became Honorary Doctor of Tartu University, Estonia.
讲座概要
If there is a model including parameters and an estimation function, e.g., the least squares function or negative log-likelihood, so called penalized estimators can be obtained. This means that the estimation function is modified by adding some "penalizing" term. In the presentation, inequality restrictions are put on the parameters (prior information) instead of putting restrictions on the estimation function which in turn also will lead to a penalized estimation function. The basic idea is to utilize convex optimization. One key result is that everything depends on the observed data if it is of advantage to penalize.
