Keynote Speakers

Dietrich von Rosen finished his Ph.D. in Mathematical Statistics at Stockholm University 1986 and stayed at the university as assistant professor until 1990. From that year he moved to Uppsala University and became a lecturer in Mathematical Statistics. The year 1998 he got a full professorship in Statistics at the Swedish University of Agricultural Sciences and from 2009 he is also working as adjoined professor in Mathematical Statistics 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 100 peer reviewed articles have been written. von Rosen has one published book entitled “Advanced Multivariate Statistics with Matrices” and a new book “Bilinear Regression Analysis, an Introduction” to be published next year. He is an associate editor of 4 international journals and most time is spent with Journal of Multivariate Analysis. At international conferences von Rosen has participated several times as an invited speaker and also organized several international conferences and workshops. In 2014 von Rosen became Honorary Doctor of Tartu University, Estonia. von Rosen has ongoing collaborations with researchers in China, India, Estonia, Poland, Mexico and the USA, among others.

Speech Title: "Safety Belt Regression"

Abstract: Penalized estimation methods will be considered (Ridge estimation). If there is a model including parameters and an estimation function, e.g., the least squares function, so called penalized estimates can be obtained. This means that the estimation function is modified by adding some penalizing term. However, in this presentation of the subject, restrictions are put on the parameters instead of the estimation function which in turn also will lead to a penalized estimation function. Indeed the two different approaches are very similar. The results will often be the same but interpretations can differ. In some way, from a likelihood point of view, it is more natural to put restrictions on the parameters in a model than on the estimation function. Moreover the presented approach can guide when ordinary estimates such as least squares or maximum likelihood estimates should be used or when penalized estimates are advantageous. Our approach is based on convex optimization theory.

Dr. Viktor Witkovsky
Slovak Academy of Sciences, Slovakia