Extreme returns in a shortfall risk framework

摘要: One of the most important aspects in asset allocation problems is assumed probability distribution returns. Financial managers generally suppose normal distribution, even if extreme realizations usually have an higher frequency than results case normally distributed Using Monte Carlo simulation, we propose and solve problem with shortfall constraint, evaluating exact risk-level for misspecification tails behaviour. In particular, optimisation problem, assume that returns are generated by a multivariate Student-t, when reality come from where each marginal Student-t different degrees freedom; this method permits us to value effective risk managers. analysed, it also interesting observe density distributions produces return level can be approximated adequately assuming problem. The present approach could instrument investors who require qualitative assessment reliability sensitivity their investment strategies models potentially misspecified. * GRETA (Venice). ** (Venice) University Venice. Extreme Returns Shortfall Risk Framework