Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Theorem 2 — lim sup⁡∣Sn∣/2Dn2log⁡log⁡Dn2≤1\limsup |S_n|/\sqrt{2D_n^2\log\log D_n^2}\le 1limsup∣Sn​∣/2Dn2​loglogDn2​​≤1 a.s. for class [G]

Proved
AzumaWeightedSums.IteratedLog.theorem2_limsup_le_one

by mikedeng1 · Oct 4, 2026 · Mathlib 0df444a (Lean v4.33.1)

law-of-the-iterated-logarithmmartingalep2o-batch-pfp1bp2o-gran-per-chapterp2o-plan-paperp2o-v1probability

Let (xn)n≥1(x_n)_{n\ge1}(xn​)n≥1​ be a sequence of random variables for which [G] holds with τ(xn)≤1\tau(x_n)\le1τ(xn​)≤1 for all nnn: martingale differences with respect to a filtration (An)(\mathfrak A_n)(An​) with E{exp⁡(txn)∣An−1}≤exp⁡(t2/2)E\{\exp(tx_n)\mid\mathfrak A_{n-1}\}\le\exp(t^2/2)E{exp(txn​)∣An−1​}≤exp(t2/2) a.s. for every real ttt. Let (an)(a_n)(an​) be a real sequence, and put Dn2=∑j=1naj2D_n^2=\sum_{j=1}^n a_j^2Dn2​=∑j=1n​aj2​ and Sn=a1x1+⋯+anxnS_n=a_1x_1+\dots+a_nx_nSn​=a1​x1​+⋯+an​xn​. If

an2/Dn2→0,Dn2→∞(n→∞),a_n^2/D_n^2\to0,\qquad D_n^2\to\infty\qquad(n\to\infty),an2​/Dn2​→0,Dn2​→∞(n→∞),

then

lim sup⁡n→∞∣Sn∣2Dn2log⁡log⁡Dn2≤1a.s.\limsup_{n\to\infty}\frac{|S_n|}{\sqrt{2D_n^2\log\log D_n^2}}\le1\quad\text{a.s.}n→∞limsup​2Dn2​loglogDn2​​∣Sn​∣​≤1a.s.

This is the upper half of the law of the iterated logarithm for weighted sums of conditionally sub-Gaussian martingale differences, including bounded martingale differences with ∣xn∣≤1|x_n|\le1∣xn​∣≤1. The constant 111 matches the classical Hartman–Wintner law for independent standard Gaussian increments, so it cannot be improved in general.

Formalization Note The lim sup is stated in its equivalent eventual form, avoiding Lean's real-valued limsup (which is 000 on unbounded sequences): for every ε>0\varepsilon>0ε>0, almost surely, ∣Sn∣≤(1+ε)2Dn2log⁡log⁡Dn2|S_n|\le(1+\varepsilon)\sqrt{2D_n^2\log\log D_n^2}∣Sn​∣≤(1+ε)2Dn2​loglogDn2​​ for all sufficiently large nnn. Since Dn2→∞D_n^2\to\inftyDn2​→∞, log⁡log⁡Dn2>0\log\log D_n^2>0loglogDn2​>0 for all large nnn, so only meaningful values of the iterated logarithm enter. Both parts of (4.1) are hypotheses. Indices start at 111; the filtration's initial σ\sigmaσ-field is arbitrary.

Preamble
import Mathlib
import Definitions.Def_AzumaWeightedSums_IteratedLog_ClassG
import Definitions.Def_AzumaWeightedSums_IteratedLog_WeightedSums

open MeasureTheory ProbabilityTheory Filter Topology
Formal statement
namespace AzumaWeightedSums.IteratedLog

/-- Azuma 1967, Theorem 2, display (4.2), p. 362: if `(x_n)` satisfies [G] with
`τ(x_n) ≤ 1` for all `n`, and the real weights `(a_n)` satisfy (4.1),
`a_n²/D_n² → 0` and `D_n² → ∞`, then
`limsup_{n→∞} |S_n| / √(2 D_n² log log D_n²) ≤ 1` almost surely.
The lim sup is stated in its equivalent eventual form: for every `ε > 0`, almost surely,
`|S_n| ≤ (1 + ε) √(2 D_n² log log D_n²)` for all sufficiently large `n`. -/
theorem theorem2_limsup_le_one {Ω : Type*} {mΩ : MeasurableSpace Ω} {μ : Measure Ω}
    [IsProbabilityMeasure μ] {ℱ : Filtration ℕ mΩ} {x : ℕ → Ω → ℝ}
    (hx : IsCondSubgaussianOne μ ℱ x) (a : ℕ → ℝ)
    (ha_ratio : Tendsto (fun n => a n ^ 2 / sqWeightSum a n) atTop (𝓝 0))
    (ha_div : Tendsto (fun n => sqWeightSum a n) atTop atTop) :
    ∀ ε : ℝ, 0 < ε → ∀ᵐ ω ∂μ, ∀ᶠ n in atTop,
      |weightedSum a x n ω|
        ≤ (1 + ε) * Real.sqrt (2 * sqWeightSum a n
            * Real.log (Real.log (sqWeightSum a n))) := by sorry

end AzumaWeightedSums.IteratedLog
Source
Azuma, Weighted sums of certain dependent random variables, Tôhoku Math. J. 19 (1967), https://doi.org/10.2748/tmj/1178243286, p. 362, Theorem 2, displays (4.1)–(4.2)
Human review
  • Endorsed by Shuze Chen · Oct 4, 2026

    Confirmed by the moderator at approval.

  • Endorsed by mikedeng1 · Oct 4, 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