A risk measurement approach from risk-averse stochastic optimization of score functions

Abstract

We propose a risk measurement approach for a risk-averse stochastic problem. We provide results that guarantee that our problem has a solution. We characterize and explore the properties of the argmin as a risk measure and the minimum as a deviation measure. We provide a connection between linear regression models and our framework. Based on this conception, we consider conditional risk and provide a connection between the minimum deviation portfolio and linear regression. Moreover, we also link the optimal replication hedging to our framework.