Risk measuring under model uncertainty (Q2428050)

The authors initially provide results related to the topological properties of a space similar to the space of integrable random variables with respect to a probability space, where a capacity defined on a vector lattice related to a function space replaces this specified probability measure. Especially, these results concern the topological dual of this space, and the most important one represents the values of any continuous linear functional of this space as integrals with respect to a regular bounded signed measure. In the next part of this article, dual representation results related to weak compactness of the previously mentioned space are proved for convex risk measures. There are also results which connect coherent and convex risk measures defined on these spaces. The last part of the article is devoted to the relation between capacities and probability measures defined on metrizable and separable spaces. There is also a part about regular risk measures, which are a sub-class of coherent risk measures defined on vector lattices related to function spaces which are also connected to the metrizable, separable space mentioned before. Examples in the context of \(G\)-expectations and uncertain volatility complete the content of the article.