397 – Section for Statistical Programmers and Analysts
Extreme Value Modeling of Minnesota River Flood
Deepak Sanjel
Minnesota State University
You-Gan Wang
University of Queensland
Accurately modeling extreme events such as flood, fire, hurricane etc. has become more and more important. Several methods of analyzing extreme values are proposed in literature however, most of them are based on the extreme value limit distributions or some related family of distributions. In this paper we review these techniques and proposed to use Bayesian method to analyze such extreme value events. These ideas are illustrated with an analysis of Minnesota river flood data. Comparisons between different models such as Block Maxima model, Picks Over Threshold (POT) model and Bayesian approach have been made.