Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Theorem (4.4) — (Z_1 + ⋯ + Z_n)/n → ν/(1 − ν) a.e. if ν = E(σ) < 1, and → ∞ a.e. if ν ≥ 1

Proved
SolomonRWRE.DiffEq.theorem_4_4

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

ergodic-theoremp2o-batch-pfp2ap2o-gran-per-chapterp2o-plan-paperp2o-v1random-difference-equationrandom-environmentstrong-law

Let σ1,σ2,…\sigma_1,\sigma_2,\dotsσ1​,σ2​,… be independent, identically distributed, nonnegative random variables, and let ZnZ_nZn​ solve the system of difference equations (4.1):

Z0=0,Zn=σn(1+Zn−1),n≥1.Z_0=0,\qquad Z_n=\sigma_n(1+Z_{n-1}),\quad n\ge1 .Z0​=0,Zn​=σn​(1+Zn−1​),n≥1.

Let ν=E(σ)∈[0,∞]\nu=E(\sigma)\in[0,\infty]ν=E(σ)∈[0,∞]. Then almost surely

lim⁡n→∞Z1+⋯+Znn={ν1−ν,ν<1,∞,ν≥1.\lim_{n\to\infty}\frac{Z_1+\dots+Z_n}{n}=\begin{cases}\dfrac{\nu}{1-\nu}, & \nu<1,\\[2mm] \infty, & \nu\ge1.\end{cases}n→∞lim​nZ1​+⋯+Zn​​=⎩⎨⎧​1−νν​,∞,​ν<1,ν≥1.​

For the random walk in a random environment, 2Zn+12Z_n+12Zn​+1 is the conditional mean first-passage time from nnn to n+1n+1n+1 given the environment (with 000 reflecting). The theorem therefore gives the growth rate of the mean time to reach level nnn in a typical environment.

Formalization Note ν\nuν is a lower Lebesgue integral in [0,∞][0,\infty][0,∞]; σ\sigmaσ need not be integrable, and ν=∞\nu=\inftyν=∞ falls in the second case. The averages are taken in [0,∞][0,\infty][0,∞], which loses nothing since Zk≥0Z_k\ge0Zk​≥0. Both cases are almost-sure limits.

Preamble
import Mathlib
import Definitions.Def_SolomonRWRE_DiffEq_System

open MeasureTheory ProbabilityTheory Filter Topology
Formal statement
namespace SolomonRWRE.DiffEq

/-- **Theorem (4.4)** (Solomon, Random Walks in a Random Environment, Ann. Probab. 3(1)
(1975), p. 28). Let `ν = E(σ)`. Then a.e.
`lim_{n→∞} (Z_1 + ⋯ + Z_n)/n = ν/(1 − ν)` if `ν < 1`, and `= ∞` if `ν ≥ 1`.

**Formalization Note.** `Z` solves (4.1) with `σ_1, σ_2, …` i.i.d. nonnegative
(`IsIIDNonneg`). `ν` is the lower Lebesgue integral of `σ_1` in `[0, ∞]` (`ν = ∞` is
allowed and falls in the second case). The averages are taken in `[0, ∞]` through
`ENNReal.ofReal`, which loses nothing as `Z_k ≥ 0`; both cases are almost-sure limits. -/
theorem theorem_4_4 {Ω : Type*} [MeasurableSpace Ω] (P : Measure Ω)
    [IsProbabilityMeasure P] (σ : ℕ → Ω → ℝ) (hσ : IsIIDNonneg P σ) :
    (nu P σ < 1 → ∀ᵐ ω ∂P,
      Tendsto (fun n : ℕ => ENNReal.ofReal ((∑ k ∈ Finset.Icc 1 n, Z σ k ω) / n)) atTop
        (𝓝 (nu P σ / (1 - nu P σ)))) ∧
    (1 ≤ nu P σ → ∀ᵐ ω ∂P,
      Tendsto (fun n : ℕ => ENNReal.ofReal ((∑ k ∈ Finset.Icc 1 n, Z σ k ω) / n)) atTop
        (𝓝 ⊤)) := by sorry

end SolomonRWRE.DiffEq
Source
Solomon, Random Walks in a Random Environment, Ann. Probab. 3(1):1–31 (1975), DOI 10.1214/aop/1176996444, p. 28, Theorem (4.4)
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