Simulation and Robust Statistics

This book addresses the topics statistical simulation and robust estimation. It is accompanied by several software packages for the open source statistical computing environment R. Simulation studies are widely used by statisticians to gain insight into the quality of developed methods. In order to facilitate the implementation of simulation experiments, a general software framework for statistical simulation has been designed. However, simulation studies in survey statistics typically require population data, which are only in exceptions available to researchers. A method for generating close-to-reality population data for complex household surveys has thus been developed. Furthermore, confidentiality issues of such simulated population data are analyzed based on different worst case scenarios. The developed simulation methodology is used in two practical applications of robust statistics. First, semiparametric methods for robust estimation of indicators on poverty and social exclusion are evaluated. Second, a robust linear model selection procedure for applications in the social sciences has been developed and is applied in the context of quality of life research.

Autorentext

received a Master's degree in Applied Mathematics and a PhD degree in Statistics from Vienna University of Technology, Austria, where he also held a research assistant position. Currently, he is a postdoctoral research fellow at Katholieke Universiteit Leuven, Belgium.