1) Justus Liebig University of Giessen (Co-ordinator)

Scientific Staff

  • Prof. Dr. Peter Winker (network coordinator)
  • Prof. Dr. Horst Rinne
  • Prof. Dr. Katja Specht
  • Prof. Dr. Wolfgang Bessler
  • Prof. Dr. Ludger Overbeck
  • Dr. Dorothea Reimer
  • Markus Spory
  • Mark Meyer
  • Vahidin Jeleskovic

Peter Winker graduated in mathematics from the University of Konstanz; he holds a Ph.D. in economics (University of Konstanz) and a habilitation in economics and econometrics (University of Mannheim); since 2001 he has been professor of economics, statistics and econometrics at IU Germany, University of Erfurt and University of Giessen.

Expertise of the Team

The team contributes expertise in the development, application and analysis of optimization heuristics, model selection, time series analysis, financial econometrics and empirical finance by members of three different departments: Department of Statistics and Econometrics (Winker, Rinne, Specht, Reimer, Spory, Meyer, and Jeleskovic), Department of Finance and Banking (Bessler) and Department of Mathematical Finance and Quantitative Risk Management (Overbeck). General aspects of optimization heuristics and their applications to problems in statistics, econometrics and finance are analyzed by Winker (2001) and Winker and Gilli (2004). A first proposal for using optimization heuristics for validating agent based models has been presented by Gilli and Winker (2003). Current research by Jeleskovic, Gilli and Winker focuses on establishing this method based on a well founded analysis of time series properties of financial markets. Meyer will work on model selection issues in the context of nonlinear time series models based on the implementations of Maringer and Winker (2004). Bessler contributes expertise in applications in empirical finance. Overbeck has contributed several publications to credit risk modelling, which is one important field of applications of the new methods (Overbeck and Schmid (2005)). Overbeck also has much experience of applied risk management and pricing techniques from his background in the financial industry.


  • M. Gilli and P. Winker: “A Global Optimization Heuristic for Estimating Agent Based Models”, Computational Statistics and Data Analysis, 2003.
  • M. Kalkbrener, H. Lotter, and L. Overbeck: “Sensible and Efficient Capital Allocation for Credit Portfolios”, Credit Risk, 2004.
  • L. Overbeck and W. Schmidt: “Modeling Default Dependence with Threshold Models”, Journal of Derivatives, 2005.
  • P. Winker and M. Gilli: “Applications of Optimization Heuristics to Estimation and Modelling Problems”, Computational Statistics and Data Analysis, 2004.

Expertise in Training Young Scientists

The team members at the Universität Giessen trained more than 10 Ph.D. students over the last five years. Currently, Winker is supervising four Ph.D. students. He also supervised Dietmar Maringer (now CCFEA, University of Essex) for his habilitation. Overbeck supervises two Ph.D. students in collaboration with industry partners.

Links within the Network

Winker and Gilli have engaged in joint research and training activities since 2001. Part of their joint research is supported by the German Research Foundation and the Swiss National Science Foundation. They also co-operated as guest editors of Computational Statistics and Data Analysis in 2003 and 2006. Peter Winker was supervisor of the habilitation thesis of Dr. Maringer who is now lecturer in computational finance and economic agents at CCFEA, University of Essex. Winker and Maringer pursue joint research on the application of optimization heuristics to estimation and modelling problems and to complex optimization problems in finance. Winker, Gilli and Kontoghiorghes are involved in joint research activities on model selection. Overbeck conducts joint research with Kalkbrener in the field of risk analysis and portfolio optimization.

Role of the Research Team

The research team will concentrate on providing a methodological framework for the assessment of the (relative) performance of heuristic optimization tools in applications to problems from statistics, econometrics and finance. In particular, further results on the empirical and theoretical convergence properties will be derived. The results will be used to establish a standard for the analysis of simultaneous convergence of optimization tools and
estimators in a stochastic setting. The researchers will also analyse the impact of optimized model selection on the estimation of nonlinear time series models with a specific focus on applications to financial market time series. Furthermore, the analysis of optimization heuristics for the evaluation of agent based models will be pursued. The team members will also provide expertise and research input in applied financial mathematics, in particular credit risk management and pricing of structured credit products. Especially the calibration and estimation of those models will benefit from the collaboration in the network. Additionally, the team members will contribute to the training and transfer of knowledge activities linked to their specific expertise.