Research areas: Spatial and time series models, Bayesian analysis, Monte Carlo methods, Applications in environmental modeling.
I am interested in the statistical modeling and analysis of observations that vary in space or in time or both, with a focus on environmental systems. A flexible class of models are state space models, and the development of filtering and parameter estimation methods for these models is a central topic of my research. Another topic is uncertainty quantification for deterministic simulation models which takes not only observational errors, but also model deficits and uncertainty of inputs into account. In simple cases, this can be done by introducing stochastic elements into the model. In more complex cases, e.g. in projections of the future climate under a given emission scenario, uncertainty is assessed by analyzing and combining the results of several different models.
Research areas: Causal inference; Computational statistics and Machine learning; Graphical models; High-dimensional statistics and Sparsity; Applications in biology, chemistry and medicine
My current research is centered around statistical inference in complex models. I am interested in computational, methodological and mathematical problems in statistics as well as in addressing questions from biology, chemistry or medicine. Recent examples include: algorithms and mathematical analysis of L1-regularized estimation in high-dimensional, sparse settings; causal inference and graphical modeling in large-scale problems and their application to biological systems; and protein inference in shotgun proteomics.
Research areas: Empirical processes, Entropy, High-dimensional models, l1-regularization, Probability inequalities, Sparsity, Statistical learning.
I am doing theoretical research on the relation between the complexity of a model and the difficulty of estimation. This work involves uniform probability inequalities for random variables indexed by a parameter, and various definitions of complexity (entropy). In the high-dimensional setup, the aim is to show that certain statistical methods can mimic an oracle that knows the optimal model. A prime example is the Lasso, which can estimate a linear model with a huge number of variables almost as well as if it were known which variables are redundant. The situation is called sparse if there are indeed only a few relevant variables. Adjusting to unknown sparsity is a challenging topic in a broad area of statistics.
Research areas: Causal inference, Graphical models, High-dimensional statistics, Interval censored data, Nonstandard asymptotics, Applications in biology and medicine.
Most of my current work consists of developing methods to learn cause-effect relationships from observational data. I aim to develop such methods for high-dimensional settings where the number of variables can be much larger than the sample size, and in which some of the variables may be unobservable. In a second line of work, I consider nonparametric estimation of survival curves based on interval censored data. In particular, I study computational, theoretical and asymptotic properties of the nonparametric maximum likelihood estimator.
Research areas: Robust Statistics; Models for Censored Data; Data Analysis, Strategies and Topics in Regression; R Software for Regression; Chemical Mass Balance and Linear Unmixing; Applications of the lognormal distribution; Applied Statistics.
Most of my research is oriented towards applications of statistical methods. This includes refining procedures to support efficient data analysis, and studying strategies to deal with widespread statistical tasks like model development. On the more theoretical side, I am involved in refining robust regression methods and developing a robust procedure in spatial statistics.
Research areas: Statistical Computing, R project, Sparse Matrices, Robustness, Clustering, Curve Estimation, Numerical Accuracy.
I have been a founding member of the R Core team and been interested in different areas of statistical computing, notably numerical accuracy and software quality assessment. Robust modelling, curve estimation, clustering algorithms and (sparse) matrices, notably for mixed effect models, are areas of
particular interest to me.
Research areas: Causal inference, Graphical models and Applied statistics.
My main research interest is the question of whether (and if yes under which circumstances) it is possible to infer causal effects from purely observational data. Apart from that I'm interested in applied problems where statistics can make a contribution to solving real world problems.
Diese Website wird in älteren Versionen von Netscape ohne graphische Elemente dargestellt. Die Funktionalität der Website ist aber trotzdem gewährleistet. Wenn Sie diese Website regelmässig benutzen, empfehlen wir Ihnen, auf Ihrem Computer einen aktuellen Browser zu installieren. Weitere Informationen finden Sie auf
The content in this site is accessible to any browser or Internet device, however, some graphics will display correctly only in the newer versions of Netscape. To get the most out of our site we suggest you upgrade to a newer browser.