Every represented risk measure is coherent
ProvedRepr.repr_coherentoperations-researchprobability
A worst case over generalized scenarios is always a coherent measure of risk. This is the constructive half of the central theorem of Artzner et al.: representability implies all four axioms. Each axiom traces to a different part of the definition — subadditivity and positive homogeneity to the maximum-of-linear form, translation invariance to the weights summing to one, monotonicity to their nonnegativity.
The converse — that every coherent measure admits such a representation — is the deeper half and is stated separately as an open problem.
As throughout CoherentRisk, the state space is finite and carries no measure of its own; a scenario is a weighting supplied as data, not derived from a probability space.
Preamble
import Definitions.Def_CoherentRiskRepresentation open CoherentRisk Repr
Formal statement
namespace Repr
theorem repr_coherent {n : ℕ} {ι : Type} [Fintype ι] [Nonempty ι]
{rho : (Fin (n+1) → ℝ) → ℝ} {P : ι → (Fin (n+1) → ℝ)}
(h : RepresentedBy rho P) : Coherent rho := by
sorry
end ReprSource
P. Artzner, F. Delbaen, J.-M. Eber and D. Heath, Coherent Measures of Risk, Mathematical Finance 9 (1999) 203-228, Section 3