Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Lemma 1 — Margin monotonicity and revenue geometry

Proved
FBMSP.lemma1_pricing_geometry

by mikeh · Oct 30, 2026 · Mathlib 0df444a (Lean v4.33.1)

market-segmentationoperations-researchpricing

For general integrable mean-zero error, let pϵ(x)p_\epsilon(x)pϵ​(x) be the least optimal price, θϵ(x)=pϵ(x)−x\theta_\epsilon(x)=p_\epsilon(x)-xθϵ​(x)=pϵ​(x)−x, and Rϵ(x)R_\epsilon(x)Rϵ​(x) the optimal customer revenue. The margin is nonincreasing on positive predicted values. If 0<x1<x20<x_1<x_20<x1​<x2​, then

S(θϵ(x1))(x2−x1)≤Rϵ(x2)−Rϵ(x1)≤S(θϵ(x2))(x2−x1).S(\theta_\epsilon(x_1))(x_2-x_1)\le R_\epsilon(x_2)-R_\epsilon(x_1)\le S(\theta_\epsilon(x_2))(x_2-x_1).S(θϵ​(x1​))(x2​−x1​)≤Rϵ​(x2​)−Rϵ​(x1​)≤S(θϵ​(x2​))(x2​−x1​).

Moreover, revenue is convex, continuous, nondecreasing, and 1-Lipschitz on that domain. No MHR assumption is imposed.

Formalization note. Weak monotonicity is intended. The 1-Lipschitz conclusion follows from the source bounds and 0≤S≤10\le S\le10≤S≤1. The optional derivative formula at continuity points is outside this target.

Preamble
import Definitions.Def_FBMSP_Model
Formal statement
namespace FBMSP

/-- Lemma 1(a)--(c), with the 1-Lipschitz consequence of its secant bounds.
The optional derivative-at-continuity-points refinement is not asserted in
this target.  Draft statement; proof remains open. -/
theorem lemma1_pricing_geometry (η : MeasureTheory.Measure ℝ)
    (hnoise : ValidNoise η) :
    AntitoneOn (margin η) (Set.Ioi (0 : ℝ)) ∧
    (∀ x₁ x₂ : ℝ, 0 < x₁ → x₁ < x₂ →
      survival η (margin η x₁) * (x₂ - x₁) ≤
          singleRevenue η x₂ - singleRevenue η x₁ ∧
        singleRevenue η x₂ - singleRevenue η x₁ ≤
          survival η (margin η x₂) * (x₂ - x₁)) ∧
    ConvexOn ℝ (Set.Ioi (0 : ℝ)) (singleRevenue η) ∧
    ContinuousOn (singleRevenue η) (Set.Ioi (0 : ℝ)) ∧
    MonotoneOn (singleRevenue η) (Set.Ioi (0 : ℝ)) ∧
    LipschitzOnWith 1 (singleRevenue η) (Set.Ioi (0 : ℝ)) := by sorry

end FBMSP
Source
Titing Cui and Michael L. Hamilton, Optimal Feature-Based Market Segmentation and Pricing, October 22, 2025 manuscript (60 PDF pages); Operations Research 74(3), 1354–1370 (2026). https://doi.org/10.1287/opre.2022.0490 ; https://ssrn.com/abstract=4151103 . Lemma 1(a)–(c), p.11; proof pp.38–39 (ec4–ec5).
Read-back

What the Lean code literally says, in plain math · gpt-6-astra

Let η\etaη be a probability measure on R\mathbb RR such that the identity is integrable and ∫e dη(e)=0\int e\,d\eta(e)=0∫edη(e)=0. Put S(t)=η([t,∞))S(t)=\eta([t,\infty))S(t)=η([t,∞)) and r(x,p)=pS(p−x)r(x,p)=pS(p-x)r(x,p)=pS(p−x). For real xxx, let Ox={p∈R:p≥0 and r(x,q)≤r(x,p) for every real q≥0}O_x=\{p\in\mathbb R:p\geq0\ \text{and}\ r(x,q)\leq r(x,p)\ \text{for every real }q\geq0\}Ox​={p∈R:p≥0 and r(x,q)≤r(x,p) for every real q≥0}, let p∗(x)=inf⁡Oxp_*(x)=\inf O_xp∗​(x)=infOx​ with the real-infimum convention inf⁡∅=0\inf\varnothing=0inf∅=0, and let R(x)=sup⁡p≥0r(x,p)R(x)=\sup_{p\geq0}r(x,p)R(x)=supp≥0​r(x,p). Define m(x)=p∗(x)−xm(x)=p_*(x)-xm(x)=p∗​(x)−x. The assertion is the conjunction of the following properties on strictly positive real xxx: if 0<x1≤x20<x_1\leq x_20<x1​≤x2​, then m(x2)≤m(x1)m(x_2)\leq m(x_1)m(x2​)≤m(x1​); whenever 0<x1<x20<x_1<x_20<x1​<x2​, one has S(m(x1))(x2−x1)≤R(x2)−R(x1)≤S(m(x2))(x2−x1)S(m(x_1))(x_2-x_1)\leq R(x_2)-R(x_1)\leq S(m(x_2))(x_2-x_1)S(m(x1​))(x2​−x1​)≤R(x2​)−R(x1​)≤S(m(x2​))(x2​−x1​); the function RRR is convex on (0,∞)(0,\infty)(0,∞), so R(λx+(1−λ)y)≤λR(x)+(1−λ)R(y)R(\lambda x+(1-\lambda)y)\leq\lambda R(x)+(1-\lambda)R(y)R(λx+(1−λ)y)≤λR(x)+(1−λ)R(y) for x,y>0x,y>0x,y>0 and 0≤λ≤10\leq\lambda\leq10≤λ≤1; RRR is continuous on (0,∞)(0,\infty)(0,∞); it is nondecreasing there, so R(x)≤R(y)R(x)\leq R(y)R(x)≤R(y) when 0<x≤y0<x\leq y0<x≤y; and it is 111-Lipschitz there, so ∣R(x)−R(y)∣≤∣x−y∣|R(x)-R(y)|\leq|x-y|∣R(x)−R(y)∣≤∣x−y∣ for x,y>0x,y>0x,y>0. The domain excludes 000 and all negative values. No density or log-concavity condition is assumed, and no derivative identity is asserted.

Human review
  • Endorsed by Shuze Chen · Oct 30, 2026

    Confirmed by the moderator at approval.

  • Endorsed by mikeh · Oct 30, 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