Network Data Analysis
Network Data are becoming more ubiquituous and describe how actors (individuals) are connected. The most simple form are social network, where individuals have friendship links. Besides this, trade networks as well as flow networks are example of network structured data. Typical research questions are whether the friend of my friend is my friend, which can be tackled with statistical tools.
Statistical Approaches for Conflict and Peace Research
We live in times of conflicts and wars and we aim to understand these events better by analysing the data traces left by conflicts. Available data is quite precise, with recorded events and remote sensing observations. These data source require sophisticated data analyses. But also sampling in and on networks is part of our current research agenda.
Quantification of Uncertainty in Machine Learning
Machine learning and AI is changing our world. We are impressed with the power of the tools, but often forget, that most models are trained on data. If training data are biased or if ambiguous, the same applies for the trained neural network. Our research focussed on the quantification of uncertainty in the era of Machine Learning and AI.
Rental Guides
"The Munich rental guide gives an overview of rents for flats in Munich and represents the local mean level of paid rents. The statistical model of the Munich rental guide 2015 is based on a representative sample with over 3000 flats in Munich. To model the mean level of rents in Munich with respect to the size, type, location, facilities and qualitiy of the flats we used a weighted general additive model. To get the upper and lower bounds of usually paid rents in Munich we used quantile regression models. As part of this project we developed the statistical model and - in cooperation with a computer scientist - a user-friendly online version of the Munich rental guide."
Coordinator(s): Prof. Dr. Goeran Kauermann, Dr. Michael Windmann
Statistical Modelling
Statistical Modelling can be seen as tool to draw information from data using advanced analysis tools. Such analyses is always driven by sustance matter research questions. Examples at our chair include efforts to...
- estimate incoming and outgoing patients from an ICU, if only the current number of admitted patients is known.
- analyse mobility based on mobile phone data.
- estimate the movement of voters between parties across elections using ecological inference.