Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Theorem 4 — Bounded loss with k segments

Proved
FBMSP.theorem4_bounded_loss

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

market-segmentationoperations-researchpricing

For any nonnegative integrable prediction law and independent integrable mean-zero error, assume finite bounds L≤M≤UL\le M\le UL≤M≤U almost surely. Then for every k≥1k\ge1k≥1,

R∞−Rk≤U−Lk.R_\infty-R_k\le\frac{U-L}{k}.R∞​−Rk​≤kU−L​.

No MHR assumption or nonnegativity of realized valuations is imposed. Prediction laws may be atomic or continuous. Taking the tight support bounds recovers the source statement; any enclosing almost-sure bounds give the same valid form.

Preamble
import Definitions.Def_FBMSP_Model
open MeasureTheory Set
Formal statement
namespace FBMSP

theorem theorem4_bounded_loss (ν η : Measure ℝ)
    (hν : ValidPredictions ν) (hη : ValidNoise η)
    (L U : ℝ) (hLU : L ≤ U) (hRange : ∀ᵐ x ∂ν, x ∈ Icc L U)
    (k : ℕ) (hk : 1 ≤ k) :
    segmentationLoss ν η k ≤ (U - L) / (k : 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 . Theorem 4, p.22; proof p.45 (ec11).
Read-back

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

Let ν,η\nu,\etaν,η be probability measures on R\mathbb RR for which the identity function is integrable, with x≥0x\geq0x≥0 for ν\nuν-almost every xxx and ∫e dη(e)=0\int e\,d\eta(e)=0∫edη(e)=0. Write S(t)=η([t,∞))S(t)=\eta([t,\infty))S(t)=η([t,∞)), r(x,p)=pS(p−x)r(x,p)=pS(p-x)r(x,p)=pS(p−x), Pν,η=∫sup⁡p≥0r(x,p) dν(x)P_{\nu,\eta}=\int\sup_{p\geq0}r(x,p)\,d\nu(x)Pν,η​=∫supp≥0​r(x,p)dν(x), and Rj(ν,η)=sup⁡P⊆[0,∞), P finite, ∣P∣≤j∫max⁡({0}∪{r(x,p):p∈P}) dν(x)R_j(\nu,\eta)=\sup_{P\subseteq[0,\infty),\,P\ {\rm finite},\,|P|\leq j}\int\max(\{0\}\cup\{r(x,p):p\in P\})\,d\nu(x)Rj​(ν,η)=supP⊆[0,∞),P finite,∣P∣≤j​∫max({0}∪{r(x,p):p∈P})dν(x). Empty menus are included and yield zero; these quantities optimize over all nonnegative real prices and do not assume any maximizing price or menu has been chosen. For all real L,UL,UL,U with L≤UL\leq UL≤U and L≤x≤UL\leq x\leq UL≤x≤U for ν\nuν-almost every xxx, and every natural k≥1k\geq1k≥1, the assertion is Pν,η−Rk(ν,η)≤(U−L)/kP_{\nu,\eta}-R_k(\nu,\eta)\leq(U-L)/kPν,η​−Rk​(ν,η)≤(U−L)/k. There is no condition L≥0L\geq0L≥0, although the assumptions require the interval to carry a probability law supported on nonnegative values and so cannot hold if U<0U<0U<0. The case L=UL=UL=U is included and gives the upper bound 000; k=0k=0k=0 is excluded. No absolute-continuity, log-concavity, bounded-shock, or nonnegative-realized-valuation condition is imposed on η\etaη.

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