p. 41, proof of Proposition 3.1, used on p. 46 — E_ℚ[ℓ(Y) | (X,Y) ∈ S] ≤ sup_y ℓ(y)
ProvedCondPortOT.BigRadius.condExp_le_supconditional-expectationdistributionally-robust-optimizationp2o-batch-pfp2ap2o-gran-per-chapterp2o-plan-paperp2o-v1
Let be a probability measure on , a measurable set with , and any function. Then
where the supremum is taken in the extended reals (it may be ) and the conditional expectation is .
This is the upper half of the uninformative-solution results: a conditional expectation of a function of can never exceed the supremum of that function over the support .
Formalization Note No measurability or boundedness of is assumed; the integrals are lower Lebesgue integrals, so is under the conventions of extIntegral.
Preamble
import Mathlib import Definitions.Def_CondPortOT_BigRadius_Setting open MeasureTheory RWPI.SqrtLasso ModelRiskOT.Duality
Formal statement
namespace CondPortOT.BigRadius
theorem condExp_le_sup {X Y : Type*} [MeasurableSpace X] [MeasurableSpace Y]
(Q : Measure (X × Y)) [IsProbabilityMeasure Q] (S : Set (X × Y)) (hS : MeasurableSet S)
(hQS : 0 < Q S) (L : Y → ℝ) :
CondPortOT.SingUninf.condExp Q S L ≤ ⨆ y, (L y : EReal) := by sorry
end CondPortOT.BigRadius
Source
Nguyen et al., Robustifying conditional portfolio decisions via optimal transport, arXiv:2103.16451v3, p. 41, proof of Proposition 3.1 (last display); invoked in the proof of Proposition 4.2, p. 46
Human review
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.