Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Corollary 1 (i) — P(Xn∈A)≤P(Yn∈A)P(X_n \in A) \le P(Y_n \in A)P(Xn​∈A)≤P(Yn​∈A) for increasing AAA

Proved
StochIneqPO.Comparison.corollary1_i

by mikedeng1 · Oct 6, 2026 · Mathlib 0df444a (Lean v4.33.1)

comparison-theoremp2o-batch-pfp2bp2o-gran-per-chapterp2o-plan-paperp2o-v1random-sequencesstochastic-order

Let EEE be a partially ordered Polish space and E1=E2=⋯=EE_1 = E_2 = \cdots = EE1​=E2​=⋯=E. Let P1,Q1P_1, Q_1P1​,Q1​ be probability measures on EEE with P1≺Q1P_1 \prec Q_1P1​≺Q1​, and for n≥2n \ge 2n≥2 let pn,qnp_n, q_npn​,qn​ be stochastic kernels from En−1E^{n-1}En−1 to EEE such that

pn(xn−1,⋅)≺qn(yn−1,⋅)whenever xn−1≤yn−1.(4)p_n(x^{n-1}, \cdot) \prec q_n(y^{n-1}, \cdot) \qquad \text{whenever } x^{n-1} \le y^{n-1}. \tag{4}pn​(xn−1,⋅)≺qn​(yn−1,⋅)whenever xn−1≤yn−1.(4)

Let (Xn)n≥1(X_n)_{n \ge 1}(Xn​)n≥1​ be a random sequence with X1∼P1X_1 \sim P_1X1​∼P1​ and conditional law pn(Xn−1,⋅)p_n(X^{n-1}, \cdot)pn​(Xn−1,⋅) of XnX_nXn​ given Xn−1X^{n-1}Xn−1, and let (Yn)(Y_n)(Yn​) be defined likewise from Q1Q_1Q1​ and (qn)(q_n)(qn​). Then for every increasing measurable set A⊆EA \subseteq EA⊆E and every nnn,

P(Xn∈A)≤P(Yn∈A).P(X_n \in A) \le P(Y_n \in A).P(Xn​∈A)≤P(Yn​∈A).

Formalization Note Both sides depend only on the law of the sequence, which is the Ionescu-Tulcea measure Kernel.trajMeasure P₁ p on ENE^{\mathbb N}EN (resp. trajMeasure Q₁ q); the statement is formulated for these laws, which covers "any random sequences" with the given initial and conditional distributions. Indices are 0-based: the paper's XnX_nXn​ is coordinate n - 1, and the paper's kernel pnp_npn​ is p (n - 2). Hypothesis (4) is pointwise for all ordered pairs of histories.

Preamble
import Mathlib
import Definitions.Def_StochIneqPO_Comparison_StochLE
Formal statement
namespace StochIneqPO.Comparison

open MeasureTheory ProbabilityTheory Finset

theorem corollary1_i {E : Type*} [TopologicalSpace E] [PolishSpace E] [MeasurableSpace E]
    [BorelSpace E] [PartialOrder E] [OrderClosedTopology E]
    (P₁ Q₁ : Measure E) [IsProbabilityMeasure P₁] [IsProbabilityMeasure Q₁]
    (p q : (n : ℕ) → Kernel (Π _ : Iic n, E) E)
    [∀ n, IsMarkovKernel (p n)] [∀ n, IsMarkovKernel (q n)]
    (hPQ : StochLE P₁ Q₁)
    (hpq : ∀ n, ∀ x y : Π _ : Iic n, E, x ≤ y → StochLE (p n x) (q n y))
    (A : Set E) (hA : MeasurableSet A) (hAup : IsUpperSet A) (n : ℕ) :
    Kernel.trajMeasure (X := fun _ => E) P₁ p {x | x n ∈ A} ≤
      Kernel.trajMeasure (X := fun _ => E) Q₁ q {x | x n ∈ A} := by sorry

end StochIneqPO.Comparison
Source
Kamae, Krengel, O'Brien, Stochastic Inequalities on Partially Ordered Spaces, Ann. Probab. 5 (1977), Corollary 1 (i), p. 904 (PDF p. 6), with the paragraph preceding it and Theorem 2, p. 903
Human review
  • Endorsed by Shuze Chen · Oct 6, 2026

    Confirmed by the moderator at approval.

  • Endorsed by mikedeng1 · Oct 6, 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