In statistics our methodological goal is to extend existing methods develop statistical tools for statistical analysis, especially for cases with high-dimensional data.
To tackle such problems we will utilize methods coming from related fields in computer science, such as machine learning and pattern recognition. We will continue our work within the area of measurement error modelling (ME). But important extensions are functional data analysis (FDA) and Boosting where the development of the methodology is far from complete.
Our research is also concerned with extending the analysis of longitudinal data. We want to include models that capture features that are nearly impossible to model explicitly.
In epidemiology our main challenge is to exploit existing epidemiological data bases including biobanks in order to assess basic causal mechanisms regarding chronic non-infectious disorders.
The collaboration with molecular biologists and geneticists is a sine qua none for future growth of epidemiology.