Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Lemma 10.2, p. 28 — if X_n ≥ 0 has mean μ_n ∈ (0,∞) and variance σ_n² = o(μ_n²), then X_n/μ_n concentrates to 1

Proved
RecMarkovGraph.IndepLB.lemma_10_2

by mikedeng1 · Oct 24, 2026 · Mathlib 0df444a (Lean v4.33.1)

chebyshevconcentrationp2o-batch-pfp2ap2o-gran-per-chapterp2o-plan-paperp2o-strengthenedp2o-v1

Let X1,X2,…X_1,X_2,\dotsX1​,X2​,… be random variables on one probability space with Xn≥0X_n\ge0Xn​≥0 almost surely and mean μn=EXn∈(0,∞)\mu_n=\mathbb E X_n\in(0,\infty)μn​=EXn​∈(0,∞), and with variance σn2\sigma_n^2σn2​ finite for all large nnn. If σn2=o(μn2)\sigma_n^2=o(\mu_n^2)σn2​=o(μn2​) as n→∞n\to\inftyn→∞, then Xn/μnX_n/\mu_nXn​/μn​ concentrates to 111: for every ε>0\varepsilon>0ε>0,

lim⁡n→∞P{∣Xnμn−1∣≤ε}=1.\lim_{n\to\infty}\mathbb P\Bigl\{\Bigl|\frac{X_n}{\mu_n}-1\Bigr|\le\varepsilon\Bigr\}=1 .n→∞lim​P{​μn​Xn​​−1​≤ε}=1.

This is the step from the second-moment ratio of Lemma 10.4 to convergence in probability of Sn/ESnS_n/\mathbb E S_nSn​/ESn​.

Differences from the source:

  • generalization: "the variance σ_n² of X_n" with σ_n² = o(μ_n²), so σ_n² finite for all large n; random variables → MemLp (X n) 2 P for all large n (∀ᶠ n in atTop) and Measurable (X n); σ_n² = ProbabilityTheory.variance (X n) P; reason: σ_n² = o(μ_n²) forces a finite variance only eventually, and the conclusion is a limit; at the finitely many n with infinite variance Lean's variance is the junk value 0, which an eventual o-condition ignores; specializes to the source, where every σ_n² may be finite
  • encoding: μ_n ∈ (0, ∞); σ_n² = o(μ_n²) → 0 < ∫ X n ∂P; (fun n => variance (X n) P) =o[atTop] (fun n => (∫ X n ∂P)^2); reason: the mean of an integrable variable is finite; Landau o as Asymptotics.IsLittleO
  • encoding: ξ_n concentrates to c: P{|ξ_n − c| ⩽ ε} → 1 for every ε > 0 (p. 4) → ∀ ε > 0, Tendsto (fun n => P{|ξ_n − c| ≤ ε}) atTop (𝓝 1); reason: the paper's own definition, with ≤ ε
Preamble
import Mathlib
import Definitions.Def_RecMarkovGraph_IndepLB_Model
Formal statement
namespace RecMarkovGraph.IndepLB

open MeasureTheory ProbabilityTheory Filter Topology

theorem lemma_10_2 {Ω : Type*} [MeasurableSpace Ω] (P : Measure Ω) [IsProbabilityMeasure P]
    (X : ℕ → Ω → ℝ) (hXm : ∀ n, Measurable (X n)) (hX0 : ∀ n, 0 ≤ᵐ[P] X n)
    (hX2 : ∀ᶠ n in atTop, MemLp (X n) 2 P) (hμ : ∀ n, 0 < ∫ ω, X n ω ∂P)
    (hσ : (fun n => variance (X n) P) =o[atTop] (fun n => (∫ ω, X n ω ∂P) ^ 2)) :
    ∀ ε : ℝ, 0 < ε →
      Tendsto (fun n => P.real {ω | |X n ω / ∫ ω', X n ω' ∂P - 1| ≤ ε}) atTop (𝓝 1) := by sorry

end RecMarkovGraph.IndepLB
Source
Gupte, Zhu, Large independent sets in recursive Markov random graphs, arXiv:2207.04514v4, p. 28, Lemma 10.2
Human review
  • Endorsed by Shuze Chen · Oct 24, 2026

    Confirmed by the moderator at approval.

  • Endorsed by mikedeng1 · Oct 24, 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