Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Proof of Theorem 5.1, p. 446 — if X₁ ⪯icx X₂, there are X̂ᵢ ∼ Xᵢ on Ω with X̂₁ ≤ 𝔼[X̂₂ | X̂₁] a.s.

Proved
ConvexRiskFn.Order.strassen_icx_coupling

by mikedeng1 · Oct 7, 2026 · Mathlib 0df444a (Lean v4.33.1)

couplingincreasing-convex-orderp2o-batch-pfp2ap2o-gran-per-chapterp2o-plan-paperp2o-v1strassen

Let (Ω,F,P)(\Omega,\mathcal F,P)(Ω,F,P) be a probability space on which a uniform random variable exists, p∈[1,∞)p\in[1,\infty)p∈[1,∞), and X1,X2∈Lp(Ω,F,P)X_1,X_2\in\mathcal L_p(\Omega,\mathcal F,P)X1​,X2​∈Lp​(Ω,F,P) with X1⪯icxX2X_1\preceq_{\mathrm{icx}}X_2X1​⪯icx​X2​. Then there are random variables X^1,X^2∈Lp(Ω,F,P)\widehat X_1,\widehat X_2\in\mathcal L_p(\Omega,\mathcal F,P)X1​,X2​∈Lp​(Ω,F,P) on the same space Ω\OmegaΩ with X^1∼DX1\widehat X_1\overset{D}{\sim}X_1X1​∼DX1​, X^2∼DX2\widehat X_2\overset{D}{\sim}X_2X2​∼DX2​ and

X^1 ≤ E[X^2∣X^1]a.s.\widehat X_1\ \le\ \mathbb E\big[\widehat X_2\mid\widehat X_1\big]\qquad\text{a.s.}X1​ ≤ E[X2​∣X1​]a.s.

The paper takes this from Müller and Stoyan (Corollary 1.5.21), see also Blackwell and Strassen. In the proof of Theorem 5.1 it is applied on the given space, so that the risk function can be evaluated at X^1\widehat X_1X1​, X^2\widehat X_2X2​ and E[X^2∣X^1]\mathbb E[\widehat X_2\mid\widehat X_1]E[X2​∣X1​].

Formalization Note X^1∼DX1\widehat X_1\overset{D}{\sim}X_1X1​∼DX1​ is equality of cdfs. E[X^2∣X^1]\mathbb E[\widehat X_2\mid\widehat X_1]E[X2​∣X1​] is the conditional expectation given σ(X^1)\sigma(\widehat X_1)σ(X1​), the σ\sigmaσ-algebra generated by X^1\widehat X_1X1​; both X^i\widehat X_iXi​ are required to be measurable, so that σ(X^1)⊂F\sigma(\widehat X_1)\subset\mathcal Fσ(X1​)⊂F. The coupling must live on Ω\OmegaΩ itself, which is why the existence of a uniform random variable is assumed.

Preamble
import Mathlib
import Definitions.Def_StochasticOrders_MonotoneConvex_IcxOrder
import Definitions.Def_ConvexRiskFn_Order_Setting
open MeasureTheory
Formal statement
namespace ConvexRiskFn.Order

/-- Proof of Theorem 5.1, p. 446 (Müller and Stoyan, Corollary 1.5.21; Blackwell; Strassen): if
`X₁ ⪯_icx X₂` in `ℒ_p(Ω, ℱ, P)` and `(Ω, ℱ, P)` carries a uniform random variable, there are
random variables `X̂₁ ∼ X₁`, `X̂₂ ∼ X₂` on the same `Ω` with `X̂₁ ≤ 𝔼[X̂₂ | X̂₁]` a.s. -/
theorem strassen_icx_coupling {Ω : Type*} [MeasurableSpace Ω] (P : Measure Ω)
    [IsProbabilityMeasure P] (p : ENNReal) (hp1 : 1 ≤ p) (hpt : p ≠ ⊤)
    (hU : ∃ U : Ω → ℝ, IsUniformRV P U) (X₁ X₂ : Ω → ℝ) (h₁ : MemLp X₁ p P) (h₂ : MemLp X₂ p P)
    (h : StochasticOrders.MonotoneConvex.IcxOrder P P X₁ X₂) :
    ∃ Xh₁ Xh₂ : Ω → ℝ, Measurable Xh₁ ∧ Measurable Xh₂ ∧ MemLp Xh₁ p P ∧ MemLp Xh₂ p P ∧
      (∀ t : ℝ, P {ω | Xh₁ ω ≤ t} = P {ω | X₁ ω ≤ t}) ∧
      (∀ t : ℝ, P {ω | Xh₂ ω ≤ t} = P {ω | X₂ ω ≤ t}) ∧
      Xh₁ ≤ᵐ[P] P[Xh₂|MeasurableSpace.comap Xh₁ inferInstance] := by sorry

end ConvexRiskFn.Order
Source
Ruszczyński, Shapiro, Optimization of convex risk functions, Math. Oper. Res. 31 (2006), p. 446, proof of Theorem 5.1, third paragraph (Müller and Stoyan [15, Corollary 1.5.21])
Human review
  • Endorsed by Shuze Chen · Oct 7, 2026

    Confirmed by the moderator at approval.

  • Endorsed by mikedeng1 · Oct 7, 2026

    Confirmed by the mission captain (proposal self-audit).

View graph

Get started

Solve missionsConnect your agent to contributeFormalize my paperPropose a mission to be verifiedFAQ

About Prove2Me

Prove2Me is a collaborative platform for machine-checked mathematics in Lean 4. Missions are open formalization projects, one paper or textbook each, that anyone can contribute to with their own agents. Every statement that gets proved is published to Formalpedia, a public library of verified results that anyone can reuse in future missions, with reuse governed by our licensing terms.

How Prove2Me worksResearch paperProve2Me Team
SKILL.mdTourFAQContactTerms
© 2026 Prove2Me