Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Example 8.1.6 — the maximum of the walk converges to the maximum of Brownian motion

Proved
DurrettProbability.donsker_max

by naimengye · Sep 19, 2026 · Mathlib 0df444a (Lean v4.33.1)

analysisprobabilitystochastic-systems

Under the hypotheses of Durrett's Theorem 8.1.2 — X0,X1,…X_0,X_1,\dotsX0​,X1​,… measurable, jointly independent, identically distributed, with mean 000 and variance 111 — and with BBB a Brownian motion all of whose paths are continuous,

max⁡0≤m≤nSmn ⟹ max⁡0≤t≤1Bt,\max_{0\le m\le n}\frac{S_m}{\sqrt n}\ \Longrightarrow\ \max_{0\le t\le1}B_t ,0≤m≤nmax​n​Sm​​ ⟹ 0≤t≤1max​Bt​,

where Sm=X0+⋯+Xm−1S_m=X_0+\dots+X_{m-1}Sm​=X0​+⋯+Xm−1​, so that the maximum on the left includes S0=0S_0=0S0​=0, and the right side is the supremum of the Brownian path over the unit interval.

Formalization Note The left side is the maximum of Sm/nS_m/\sqrt nSm​/n​ over the n+1n+1n+1 indices 0≤m≤n0\le m\le n0≤m≤n, taken as a maximum over a non-empty finite set. At n=0n=0n=0 the rescaling divides by 0=0\sqrt0=00​=0, which in the reals is 000; this has no effect on a limit along n→∞n\to\inftyn→∞.

The right side is the supremum of t↦Bt(ω)t\mapsto B_t(\omega)t↦Bt​(ω) over [0,1][0,1][0,1], written as a supremum in the reals. Because the path is continuous and the interval is compact and non-empty, this supremum is attained and is the maximum the book writes; no separate boundedness hypothesis is needed.

Both sides are non-negative, since m=0m=0m=0 and t=0t=0t=0 contribute 000. Durrett's companion formula P(M1≥a)=2 P(B1≥a)\mathbb P(M_1\ge a)=2\,\mathbb P(B_1\ge a)P(M1​≥a)=2P(B1​≥a), which identifies the limit law through the reflection principle, is not part of this statement.

Preamble
import Mathlib
import Definitions.Def_DurrettProbability_Brownian
import Definitions.Def_DurrettProbability_Donsker

open Filter MeasureTheory ProbabilityTheory
open scoped NNReal Topology
Formal statement
namespace DurrettProbability

theorem donsker_max
    {Ω : Type*} [MeasurableSpace Ω] {P : Measure Ω} [IsProbabilityMeasure P]
    {X : ℕ → Ω → ℝ} (hmeas : ∀ k, Measurable (X k)) (hindep : iIndepFun X P)
    (hident : ∀ k, IdentDistrib (X k) (X 0) P P) (hint : Integrable (fun ω => X 0 ω ^ 2) P)
    (hmean : ∫ ω, X 0 ω ∂P = 0) (hvar : ∫ ω, X 0 ω ^ 2 ∂P = 1)
    {Ω' : Type*} [MeasurableSpace Ω'] {P' : Measure Ω'} [IsProbabilityMeasure P']
    {B : ℝ≥0 → Ω' → ℝ} (hB : IsBrownianReal B P') (hBc : ∀ ω, Continuous fun t => B t ω) :
    TendstoInDistribution
      (fun n : ℕ => fun ω => (Finset.range (n + 1)).sup' Finset.nonempty_range_add_one
        fun m => (∑ k ∈ Finset.range m, X k ω) / Real.sqrt n)
      atTop
      (fun ω => ⨆ t : Set.Icc (0 : ℝ) 1, B ⟨t, t.2.1⟩ ω) (fun _ => P) P' := by sorry

end DurrettProbability
Source
Durrett, Probability: Theory and Examples, Version 5 (11 January 2019), p. 392 (PDF p. 400), Example 8.1.6: 'Maxima. Let psi(omega) = max{omega(t) : 0 <= t <= 1}. Again, psi : C[0,1] -> R is continuous. This time Theorem 8.1.5 implies max_{0 <= m <= n} S_m/sqrt(n) => M_1 = max_{0 <= t <= 1} B_t. To complete the picture, we observe that by (7.4.4) the distribution of the right-hand side is P_0(M_1 >= a) = P_0(T_a <= 1) = 2 P_0(B_1 >= a).' sha256 aeac36cbf5e44c53d69fa60a2d29a393e2d0e8c955ee103bd845d925fd910886
Human review
  • Endorsed by Shuze Chen · Sep 25, 2026

    Confirmed by the moderator at approval.

  • Endorsed by naimengye · Sep 25, 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