Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Lemma 9, p. 58 (corrected: R_1 ≥ 0) — for w_2 = mτ + zq_{2→1}(τ), m > R_1, and z/(m − R_1) in the window, r(u, 2, w, σ) > 0 for all u, σ_2; the window is nonempty

Proved
DriverSurge.IC.lemma_9

by mikedeng1 · Oct 24, 2026 · Mathlib 0df444a (Lean v4.33.1)

incentive-compatibilityp2o-batch-pfp2ap2o-correctedp2o-gran-per-chapterp2o-plan-paperp2o-strengthenedp2o-v1surge-pricing

Fix a non-surge policy σ1\sigma_1σ1​ and a payment w1w_1w1​ integrable on it, with Q1=Q1(σ1)Q_1=Q_1(\sigma_1)Q1​=Q1​(σ1​), T1=T1(σ1)T_1=T_1(\sigma_1)T1​=T1​(σ1​) and R1=R1(w1,σ1)≥0R_1=R_1(w_1,\sigma_1)\ge0R1​=R1​(w1​,σ1​)≥0. Let Qˉ2\bar Q_2Qˉ​2​, Tˉ2\bar T_2Tˉ2​ be the values of Q2Q_2Q2​, T2T_2T2​ at σ2=(0,∞)\sigma_2=(0,\infty)σ2​=(0,∞), assume finite means, and let w2(τ)=mτ+z q2→1(τ)w_2(\tau)=m\tau+z\,q_{2\to1}(\tau)w2​(τ)=mτ+zq2→1​(τ) with m>R1m>R_1m>R1​. If

T1(λ2→1Tˉ2−Qˉ2)−(Q1+T1λ2→1)Q1(λ2→1Tˉ2−Qˉ2)+λ2→1(Q1+T1λ2→1)<zm−R1<Qˉ2T1+Q1Q1(Qˉ2−λ2→1),\frac{T_1(\lambda_{2\to1}\bar T_2-\bar Q_2)-(Q_1+T_1\lambda_{2\to1})}{Q_1(\lambda_{2\to1}\bar T_2-\bar Q_2)+\lambda_{2\to1}(Q_1+T_1\lambda_{2\to1})}<\frac{z}{m-R_1}<\frac{\bar Q_2T_1+Q_1}{Q_1(\bar Q_2-\lambda_{2\to1})},Q1​(λ2→1​Tˉ2​−Qˉ​2​)+λ2→1​(Q1​+T1​λ2→1​)T1​(λ2→1​Tˉ2​−Qˉ​2​)−(Q1​+T1​λ2→1​)​<m−R1​z​<Q1​(Qˉ​2​−λ2→1​)Qˉ​2​T1​+Q1​​,

then r(u,2,w,σ)>0r(u,2,w,\sigma)>0r(u,2,w,σ)>0 for every u>0u>0u>0 and every open σ2⊆(0,∞)\sigma_2\subseteq(0,\infty)σ2​⊆(0,∞); hence ∂R(w,σ)/∂u>0\partial R(w,\sigma)/\partial u>0∂R(w,σ)/∂u>0 wherever f2(u)>0f_2(u)>0f2​(u)>0. Furthermore the constraint set is feasible regardless of the primitives: the lower bound is strictly below the upper bound.

Lemma 9 makes accepting every surge trip optimal; it is one half of the incentive-compatibility argument of Theorem 3.

Formalization Note Lean state 0 is the paper's non-surge state 1 and Lean state 1 the surge state 2; other i is the state j≠ij\ne ij=i. The denominators are positive: λ2→1Tˉ2−Qˉ2≥0\lambda_{2\to1}\bar T_2-\bar Q_2\ge0λ2→1​Tˉ2​−Qˉ​2​≥0 and Qˉ2>λ2→1\bar Q_2>\lambda_{2\to1}Qˉ​2​>λ2→1​ (Remark 4).

Differences from the source:

  • correction: R1R_1R1​ arbitrary → R1≥0R_1\ge0R1​≥0 assumed; reason: the proof's step "m/(m−R1)>1m/(m-R_1)>1m/(m−R1​)>1" (p. 66) needs R1≥0R_1\ge0R1​≥0, and the lemma fails without it, even under D.3.1's standing assumptions: with F1=F2=Exp(1)F_1=F_2=\mathrm{Exp}(1)F1​=F2​=Exp(1), λ1=1\lambda_1=1λ1​=1, λ2=0.1\lambda_2=0.1λ2​=0.1, λ1→2=λ2→1=1\lambda_{1\to2}=\lambda_{2\to1}=1λ1→2​=λ2→1​=1, σ1=(0,∞)\sigma_1=(0,\infty)σ1​=(0,∞), w1≡−1w_1\equiv-1w1​≡−1 (so R1=−1/2R_1=-1/2R1​=−1/2), m=0.1m=0.1m=0.1, z/(m−R1)=−0.468z/(m-R_1)=-0.468z/(m−R1​)=−0.468 inside the window (−0.935,76.5)(-0.935,76.5)(−0.935,76.5), and σ2=∅\sigma_2=\emptysetσ2​=∅, one gets r(0.001,2,w,σ)≈−0.30<0r(0.001,2,w,\sigma)\approx-0.30<0r(0.001,2,w,σ)≈−0.30<0; state 2 dominates state 1 there, since R1(w1,σ1′)≤0R_1(w_1,\sigma_1')\le0R1​(w1​,σ1′​)≤0 for every σ1′\sigma_1'σ1′​ while R2(w2,(2,∞))≈0.002>0R_2(w_2,(2,\infty))\approx0.002>0R2​(w2​,(2,∞))≈0.002>0 (checks/lemma9_counterexample_surge.py)
  • reading: "∂/∂uR(w,σ)>0\partial/\partial u R(w,\sigma)>0∂/∂uR(w,σ)>0, for all uuu" → r(u,i,w,σ)>0r(u,i,w,\sigma)>0r(u,i,w,σ)>0 for every u>0u>0u>0; reason: D.3.1 fixes ∝\propto∝ as "same sign except where f(u)=0f(u)=0f(u)=0"; the proof (p. 65) establishes positivity of the ∝\propto∝ expression
  • reading: "the constraint set is feasible" → the lower end of the window is strictly below the upper end; reason: the proof (p. 66) shows exactly this inequality
  • encoding: Tˉ2\bar T_2Tˉ2​ implicitly finite → finite means; w1w_1w1​ F1F_1F1​-integrable on σ1\sigma_1σ1​; reason: Tˉ2\bar T_2Tˉ2​, R1R_1R1​ are undefined otherwise; only undefined cases are excluded
  • generalization: D.3.1's standing assumptions (p. 56): FiF_iFi​ has a bounded density, and state 2 dominates state 1 (∃σ2\exists\sigma_2∃σ2​ with Δ(σ2,σ1)>0\Delta(\sigma_2,\sigma_1)>0Δ(σ2​,σ1​)>0 for all σ1\sigma_1σ1​) → not assumed; reason: the proof (pp. 65–66) is algebra on rrr and does not use them (with R1≥0R_1\ge0R1​≥0 the lemma held on every one of 300 random primitives × 30 random σ2\sigma_2σ2​, checks/lemma9_10_check.py), so it specializes to the source
Preamble
import Mathlib
import Definitions.Def_DriverSurge_IC_Pricing
Formal statement
namespace DriverSurge.IC

open MeasureTheory Set

/-- Lemma 9 (p. 58; corrected: `R_1 ≥ 0`). Fix a non-surge policy `s₁` (state 1 = Lean `0`) and a
payment `w_1` integrable on it, with `Q_1 = Q_1(s₁)`, `T_1 = T_1(s₁)`, `R_1 = R_1(w_1, s₁) ≥ 0`; let
`Q̄_2`, `T̄_2` be the surge-state values at `(0, ∞)`, and `w_2(τ) = mτ + zq_{2→1}(τ)` with `m > R_1`. If
`[T_1(λ_{2→1}T̄_2 − Q̄_2) − (Q_1 + T_1λ_{2→1})] / [Q_1(λ_{2→1}T̄_2 − Q̄_2) + λ_{2→1}(Q_1 + T_1λ_{2→1})] < z/(m − R_1) < (Q̄_2T_1 + Q_1)/(Q_1(Q̄_2 − λ_{2→1}))`,
then `r(u, 2, w, {s₁, s₂}) > 0` for every `u > 0` and every open `s₂ ⊆ (0, ∞)` (so `∂/∂u R(w, σ) > 0` in
the `∝` sense of D.3.1). Moreover the lower bound is below the upper bound (the constraint set is
feasible). -/
theorem lemma_9 (P : Primitives) (hmean : FiniteMean P) (w : Fin 2 → ℝ → ℝ) (s₁ : Set ℝ)
    (hs₁ : IsAdmissible s₁) (hw0 : IntegrableOn (w 0) s₁ (P.F 0)) (m z : ℝ)
    (hw1 : w 1 = fun τ => m * τ + z * q P 1 τ)
    (hR : 0 ≤ stateRate P 0 (w 0) s₁) (hm : stateRate P 0 (w 0) s₁ < m) :
    ((Tn P 0 s₁ * (P.sw 1 * Tn P 1 (Ioi 0) - Qn P 1 (Ioi 0)) - (Qn P 0 s₁ + Tn P 0 s₁ * P.sw 1)) /
          (Qn P 0 s₁ * (P.sw 1 * Tn P 1 (Ioi 0) - Qn P 1 (Ioi 0)) +
            P.sw 1 * (Qn P 0 s₁ + Tn P 0 s₁ * P.sw 1)) <
        z / (m - stateRate P 0 (w 0) s₁) →
      z / (m - stateRate P 0 (w 0) s₁) <
        (Qn P 1 (Ioi 0) * Tn P 0 s₁ + Qn P 0 s₁) / (Qn P 0 s₁ * (Qn P 1 (Ioi 0) - P.sw 1)) →
      ∀ s₂ : Set ℝ, IsAdmissible s₂ → ∀ u : ℝ, 0 < u → 0 < rFun P w ![s₁, s₂] 1 u) ∧
    (Tn P 0 s₁ * (P.sw 1 * Tn P 1 (Ioi 0) - Qn P 1 (Ioi 0)) - (Qn P 0 s₁ + Tn P 0 s₁ * P.sw 1)) /
          (Qn P 0 s₁ * (P.sw 1 * Tn P 1 (Ioi 0) - Qn P 1 (Ioi 0)) +
            P.sw 1 * (Qn P 0 s₁ + Tn P 0 s₁ * P.sw 1)) <
      (Qn P 1 (Ioi 0) * Tn P 0 s₁ + Qn P 0 s₁) / (Qn P 0 s₁ * (Qn P 1 (Ioi 0) - P.sw 1)) := by sorry

end DriverSurge.IC
Source
Garg, Nazerzadeh, Driver Surge Pricing, arXiv:1905.07544v4, p. 58, Lemma 9 (proof pp. 65–66)
Human review
  • Endorsed by Shuze Chen · Oct 24, 2026

    Confirmed by the moderator at approval.

  • Endorsed by mikedeng1 · Oct 24, 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