Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Fact 2.2 — ℋ_l ⊂ ℋ_{l−1} with ‖·‖_{ℋ_{l−1}} ≤ T‖·‖_{ℋ_l}, K_T = 0 for K ∈ ℋ_l (l > 0), ℳ_{−1}·ℋ_l ⊂ ℋ_{l−1}, and the ℳ_l inclusion

Proved
MFGLiquidation.Equilibrium.fact_2_2

by mikedeng1 · Oct 8, 2026 · Mathlib 0df444a (Lean v4.33.1)

p2o-batch-pfp2bp2o-gran-per-chapterp2o-plan-paperp2o-v1weighted-spaces

Let T>0T>0T>0, let G\mathbb GG be a filtration and l∈Rl\in\mathbb Rl∈R, and let Hl\mathcal H_lHl​, Ml\mathcal M_lMl​ be the weighted spaces of Definition 2.1. Then:

  1. Hl⊂Hl−1\mathcal H_l\subset\mathcal H_{l-1}Hl​⊂Hl−1​ and ∥Y∥Hl−1≤T ∥Y∥Hl\|Y\|_{\mathcal H_{l-1}}\le T\,\|Y\|_{\mathcal H_l}∥Y∥Hl−1​​≤T∥Y∥Hl​​ (stated for squared norms, ∥Y∥Hl−12≤T2∥Y∥Hl2\|Y\|^2_{\mathcal H_{l-1}}\le T^2\|Y\|^2_{\mathcal H_l}∥Y∥Hl−1​2​≤T2∥Y∥Hl​2​).
  2. If K∈HlK\in\mathcal H_lK∈Hl​ with l>0l>0l>0, then KT=0K_T=0KT​=0 a.s.
  3. If K1∈M−1K_1\in\mathcal M_{-1}K1​∈M−1​ has a.s. continuous paths on [0,T)[0,T)[0,T) and K2∈HlK_2\in\mathcal H_lK2​∈Hl​, then K1K2∈Hl−1K_1K_2\in\mathcal H_{l-1}K1​K2​∈Hl−1​.
  4. Ml⊂Ml−1\mathcal M_l\subset\mathcal M_{l-1}Ml​⊂Ml−1​ and ∥Y∥Ml−1≤T ∥Y∥Ml\|Y\|_{\mathcal M_{l-1}}\le T\,\|Y\|_{\mathcal M_l}∥Y∥Ml−1​​≤T∥Y∥Ml​​.

In formulas, item 3 says that

E[sup⁡0≤t≤T∣K1(t)K2(t)∣2(T−t)2(l−1)]<∞.\mathbb E\Big[\sup_{0\le t\le T}\frac{|K_1(t)K_2(t)|^2}{(T-t)^{2(l-1)}}\Big]<\infty .E[0≤t≤Tsup​(T−t)2(l−1)∣K1​(t)K2​(t)∣2​]<∞.

These facts are the bookkeeping behind the method of continuation: A∈M−1A\in\mathcal M_{-1}A∈M−1​ multiplies X∈HαX\in\mathcal H_\alphaX∈Hα​ into Hα−1\mathcal H_{\alpha-1}Hα−1​, and membership in Hl\mathcal H_lHl​ encodes the liquidation constraint.

Formalization Note The paper's last line says "the first two properties also hold for the space Ml\mathcal M_lMl​". Only the first is stated (item 4): the Ml\mathcal M_lMl​ norm is an essential supremum over dt⊗dPdt\otimes d\mathbb Pdt⊗dP, and {T}×Ω\{T\}\times\Omega{T}×Ω is null, so "K∈MlK\in\mathcal M_lK∈Ml​, l>0⇒KT=0l>0\Rightarrow K_T=0l>0⇒KT​=0" does not follow from the printed definition. Item 3 as printed (K1∈M−1K_1\in\mathcal M_{-1}K1​∈M−1​ only) is false for the same reason: the H\mathcal HH norm takes a pointwise supremum over ttt, while M−1\mathcal M_{-1}M−1​ controls K1K_1K1​ only off a dt⊗dPdt\otimes d\mathbb Pdt⊗dP-null set, on which K1K_1K1​ may be arbitrarily large. The hypothesis that K1K_1K1​ has a.s. continuous paths on [0,T)[0,T)[0,T) (true of the Riccati solution AAA in the paper) restores it. Norms are computed in [0,∞][0,\infty][0,∞] with the weight (T−t)−l(T-t)^{-l}(T−t)−l in ℝ≥0∞.

Preamble
import Mathlib
import Definitions.Def_MFGLiquidation_Equilibrium_Setting

open MeasureTheory ProbabilityTheory Filter Topology
open scoped ENNReal NNReal
Formal statement
namespace MFGLiquidation.Equilibrium

open Peng1990.SMP

theorem fact_2_2 {Ω : Type*} [MeasurableSpace Ω] (G : Filtration ℝ≥0 ‹MeasurableSpace Ω›)
    (P : Measure Ω) [SFinite P] (T : ℝ≥0) (hT : 0 < T) :
    (∀ (l : ℝ) (Y : ℝ≥0 → Ω → ℝ), MemH G P T l Y →
      MemH G P T (l - 1) Y ∧ hNormSq T P (l - 1) Y ≤ (T : ℝ≥0∞) ^ 2 * hNormSq T P l Y) ∧
    (∀ (l : ℝ) (K : ℝ≥0 → Ω → ℝ), 0 < l → MemH G P T l K → ∀ᵐ ω ∂P, K T ω = 0) ∧
    (∀ (l : ℝ) (K₁ K₂ : ℝ≥0 → Ω → ℝ), MemM G P T (-1) K₁ →
      (∀ᵐ ω ∂P, ContinuousOn (fun t => K₁ t ω) (Set.Iio T)) → MemH G P T l K₂ →
      MemH G P T (l - 1) (fun t ω => K₁ t ω * K₂ t ω)) ∧
    (∀ (l : ℝ) (Y : ℝ≥0 → Ω → ℝ), MemM G P T l Y →
      MemM G P T (l - 1) Y ∧ mNorm T P (l - 1) Y ≤ (T : ℝ≥0∞) * mNorm T P l Y) := by sorry

end MFGLiquidation.Equilibrium
Source
Fu, Graewe, Horst, Popier, A Mean Field Game of Optimal Portfolio Liquidation, arXiv:1804.04911v3, p. 8, Fact 2.2
Human review
  • Endorsed by Shuze Chen · Oct 8, 2026

    Confirmed by the moderator at approval.

  • Endorsed by mikedeng1 · Oct 8, 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