Title: Bayesian analysis of risk measures in finance
Authors: Jacinto Martin Jimenez - Universidad de Extremadura (Spain) [presenting]
Eva Lopez Sanjuan - Universidad de Extremadura (Spain)
M Isabel Parra Arevalo - Universidad de Extremadura (Spain)
Abstract: Assessing the probability of rare and extreme events is an important issue in risk management of financial portfolios. The most traditional risk measures (RM) in this context are Value at Risk, Expected Shortfall and the Return Level. Usually, a flexible Generalized Pareto Distribution (GPD) is used to model the values over a certain threshold. The parameter of this distribution have an great impact on the values of the RM. Considering a Bayesian analysis we can obtain a predicted distribution for those values. This distribution gives us more information than using a point estimation of the parametres of the GPD. We illustrate the ideas and compare both approaches with several simulation examples.