THESys Member Nadja Klein was granted with a Emmy-Noether research group in economics for excellent young researchers by the German research foundation (DFG). The Emmy Noether Programme gives exceptionally qualified early career researchers the chance to qualify for the post of professor at a university by leading an independent junior research group for a period of six years. Nadja Klein is an assistant professor of Applied Statistics at the School of Business and Economics at Humboldt-Universität zu Berlin. Together with her newly funded team of one postdoc and one doctoral student, she will develop a general framework of statistical methods and inferential techniques for a wide range of big-data applications, such as in economics (e.g. marketing) or natural and life sciences (e.g. weather forecast, chronic diseases).
The funded project will develop statistical tools for flexible high-dimensional regression models, as well as necessary scalable estimation methods. Proposing an explicit link between Bayesian probabilistic approaches and algorithms in machine learning, the research programme aims to develop more efficient statistical learning algorithms to estimate models for huge data sets. To obtain more accurate regression models for the entire distribution (instead of only the mean) simultaneously, new distributional models based on copulas and transformation models in combination with variable and shrinkage methods will be constructed for univariate as well as multivariate responses.