250430 VO Selected topics in statistics (2006S)

Statistical Extreme Value Theory deals with modeling extreme outcome of random
processes, ie. with events that happen far out in tails of a probability distribution. Its techniques have applications in areas as diverse as climate
research and meteorology, financial mathematics and risk analysis, engineering, aso. We start with an introduction to the basic concepts of Extreme Value Theory, such as the fundamental Fisher-Tippett theorem, block maxima approaches, and threshold models. Subsequently, the emphasis will lie on the application of theses techniques -- we will explore climatological data sets (with quantities such as extreme rainfall or minimal temperature) in terms of their extreme events and statistically model this behavior using the methods studied before. Moreover, we will use the statistical software "R" to better illustrate the examples.
The course uses the textbook from Stuart Coles "An introduction to Statistical Modeling of Extreme Values". Current course information during the semester can be found at http://homepage.univie.ac.at/ulrike.schneider/ExtVO.html.