Willem R. van Zwet Award won by Stéphanie van der Pas - VVSOR - VVSOR

Netherlands Society for Statistics and Operations Research | Dutch

Willem R. van Zwet Award won by Stéphanie van der Pas

This year we received five nominations for the van Zwet award. All five were high quality theses, and together they form a nice profile of our society by their variety of topics on Statistics and Operations Research. It was going to be difficult choice for the jury. Nevertheless, the jury saw two theses standing out.

Stéphanie van der Pas

The winner is Stéphanie van der Pas, title of the thesis is Topics in Mathematical and Applied Statistics, supervised by professor Aad van der Vaart from the Universiteit Leiden. She defended her thesis also in February 2017. Stephanie’s thesis is based on 7 papers that are spread over 4 topics in statistics: frequentist properties of Bayesian horseshoe prior; network analysis; sequentially collected data; and the last part is the application to survival analysis of hip replacements. All these papers found their way to international scientific journals. Specifically, the jury would like to mention chapter 3 which is based on a paper in which Stephanie shows that Bayesian methods do not correct automatically for multiplicity. This paper together with discussions of experts is published in Bayesian analysis, and has already hundred plus citations. It will likely become an influential paper.

Krzysztof Postek

The runner-up: Krzysztof Postek, title of the thesis is Distributionally and Integer Adjustable Robust Optimization, supervised by professor Dick den Hertog from the Universiteit Tilburg. He defended his thesis in February 2017. Krzysztof’s thesis is about new optimization models and algorithms that are robust with respect to all kinds of uncertainty in the data. The thesis is based on five papers, most of these are now published in high-ranked Operations Research journals. The jury was impressed by the scientific quality both theoretically and practically. And specifically the chapter where robust optimization is applied to stochastic programming problems is remarkable because it combines two major but rather different fields in OR.