(3.8), p. 6 — ρ*(·, ω) is the indicator of a closed convex 𝒜(ω) ⊂ 𝒫_{𝒴₂|ℱ₁}(ω) and ρ_ω(X) = sup_{μ∈𝒜(ω)} ⟨μ, X⟩
ProvedCondRiskMap.Repr.envelope_representation_3_8conditional-riskconvex-conjugatep2o-batch-pfp2ap2o-gran-per-chapterp2o-plan-paperp2o-v1risk-enveloperisk-mapping
Let be paired locally convex spaces over satisfying (C) and (C′), and let be a positively homogeneous, lower semicontinuous conditional risk mapping. Write . Then for every :
- is the indicator function of : it equals on and off it;
- is closed (in the topology of ) and convex;
- ;
- for every ,
the supremum being the least upper bound of the real numbers , .
The multifunction is the conditional risk envelope; (3.8) is the form of the representation on which §4 builds.
Formalization Note The supremum in (3.8) is stated with IsLUB over the image of ; this also asserts that is nonempty and the set is bounded above.
Preamble
import Mathlib import Definitions.Def_CondRiskMap_Repr_Setting
Formal statement
namespace CondRiskMap.Repr
theorem envelope_representation_3_8 {Ω : Type*} [MeasurableSpace Ω] {𝒳 𝒴 : Type*}
[AddCommGroup 𝒳] [Module ℝ 𝒳] [TopologicalSpace 𝒳]
[IsTopologicalAddGroup 𝒳] [ContinuousSMul ℝ 𝒳] [LocallyConvexSpace ℝ 𝒳]
[AddCommGroup 𝒴] [Module ℝ 𝒴] [TopologicalSpace 𝒴]
[IsTopologicalAddGroup 𝒴] [ContinuousSMul ℝ 𝒴] [LocallyConvexSpace ℝ 𝒴]
(S : PairedSpaces Ω 𝒳 𝒴) (hC : CondC S) (hC' : CondC' S)
(ρ : 𝒳 → 𝒳) (hρ : IsCondRiskMapping S ρ) (hph : PosHomogeneous ρ) (hlsc : IsLsc S ρ) :
∀ ω, (∀ μ ∈ envelope S ρ ω, conj S ρ μ ω = 0) ∧ (∀ μ ∉ envelope S ρ ω, conj S ρ μ ω = ⊤) ∧
IsClosed (envelope S ρ ω) ∧ Convex ℝ (envelope S ρ ω) ∧
envelope S ρ ω ⊆ ProbGiven S ω ∧
∀ X, IsLUB ((fun μ => pair (S.toMeasure μ) (S.toFun X)) '' envelope S ρ ω) (ρω S ρ ω X) := by sorry
end CondRiskMap.Repr
Source
Ruszczyński, Shapiro, Conditional risk mappings, preprint dated February 21, 2004, p. 6, paragraph after Proposition 1, (3.8)
Human review
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.