Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Proof of Proposition 5, eqs. (54)–(55), p. 587 — ψ₁(x) > ψ₂(x) iff π(x) > β ln K/(λ₁ − λ₂)

Proved
BoundedNV.Stakes.eq54_55_single_crossing

by mikedeng1 · Oct 6, 2026 · Mathlib 0df444a (Lean v4.33.1)

bounded-rationalitylogit-choicep2o-batch-pfp2bp2o-gran-per-chapterp2o-plan-paperp2o-v1single-crossing

Let S⊆RS\subseteq\mathbb RS⊆R have positive Lebesgue measure, let π\piπ be a real function, β>0\beta>0β>0 and λ1>λ2\lambda_1>\lambda_2λ1​>λ2​. For i=1,2i=1,2i=1,2 let ui=λiπu_i=\lambda_i\piui​=λi​π, assume eui/βe^{u_i/\beta}eui​/β is integrable on SSS, and let

ψi(x)=eui(x)/β∫Seui(v)/β dv\psi_i(x)=\frac{e^{u_i(x)/\beta}}{\int_S e^{u_i(v)/\beta}\,dv}ψi​(x)=∫S​eui​(v)/βdveui​(x)/β​

be the logit density of uiu_iui​ on SSS. Put K=∫Seu1(v)/βdv/∫Seu2(v)/βdvK=\int_S e^{u_1(v)/\beta}dv\big/\int_S e^{u_2(v)/\beta}dvK=∫S​eu1​(v)/βdv/∫S​eu2​(v)/βdv. Then for every x∈Sx\in Sx∈S,

ψ1(x)>ψ2(x)  ⟺  e(u1(x)−u2(x))/β>K  ⟺  π(x)>βln⁡Kλ1−λ2.\psi_1(x)>\psi_2(x)\iff e^{(u_1(x)-u_2(x))/\beta}>K\iff \pi(x)>\frac{\beta\ln K}{\lambda_1-\lambda_2}.ψ1​(x)>ψ2​(x)⟺e(u1​(x)−u2​(x))/β>K⟺π(x)>λ1​−λ2​βlnK​.

This single-crossing property says the decision maker with the larger stake puts more weight exactly on the decisions whose payoff exceeds one threshold; it drives the stochastic dominance (56)–(57) and hence Proposition 5.

Formalization Note β>0\beta>0β>0, positive measure of SSS and integrability of eui/βe^{u_i/\beta}eui​/β on SSS are implicit on the page; they make both normalisers finite and positive, so K>0K>0K>0 and ln⁡K\ln KlnK is genuine.

Preamble
import Mathlib
import Definitions.Def_BoundedNV_ExpFam_Logit
Formal statement
namespace BoundedNV.Stakes

open MeasureTheory

/-- Proof of Proposition 5, eqs. (54)–(55), p. 587: with `K` the ratio of the normalizers of the
logit densities `ψᵢ` of the utilities `uᵢ = λᵢ π` (`λ₁ > λ₂`, same `β`), for every `x ∈ S`,
`ψ₁(x) > ψ₂(x) ⇔ e^{(u₁(x) − u₂(x))/β} > K ⇔ π(x) > β ln K / (λ₁ − λ₂)`. -/
theorem eq54_55_single_crossing (S : Set ℝ) (hSpos : 0 < volume S)
    (π : ℝ → ℝ) (β lam₁ lam₂ : ℝ) (hβ : 0 < β) (hlam : lam₂ < lam₁)
    (hint₁ : IntegrableOn (fun x => Real.exp (lam₁ * π x / β)) S)
    (hint₂ : IntegrableOn (fun x => Real.exp (lam₂ * π x / β)) S)
    (x : ℝ) (hx : x ∈ S) :
    let K := (∫ v in S, Real.exp (lam₁ * π v / β)) / (∫ v in S, Real.exp (lam₂ * π v / β))
    (BoundedNV.Uniform.logitDensity S (fun y => lam₂ * π y) β x < BoundedNV.Uniform.logitDensity S (fun y => lam₁ * π y) β x ↔
        K < Real.exp ((lam₁ * π x - lam₂ * π x) / β)) ∧
      (K < Real.exp ((lam₁ * π x - lam₂ * π x) / β) ↔
        β * Real.log K / (lam₁ - lam₂) < π x) := by sorry

end BoundedNV.Stakes
Source
Su, Bounded Rationality in Newsvendor Models, Manufacturing & Service Operations Management 10(4), 2008, p. 587 (PDF p. 22), proof of Proposition 5, eqs. (54)–(55)
Human review
  • Endorsed by Shuze Chen · Oct 6, 2026

    Confirmed by the moderator at approval.

  • Endorsed by mikedeng1 · Oct 6, 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