Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

(32), p. 39 — the Logit diversion ratio is Σ_{k≠j} e^{−tᵏ}/(e^z + Σ_{k≠j} e^{−tᵏ}), independent of tʲ, and equals Σ_{k≠j}∂nᵏ/∂tʲ over −∂nʲ/∂tʲ

Proved
NetFeePlatform.InteropFee.diversion_logit

by mikedeng1 · Oct 14, 2026 · Mathlib 0df444a (Lean v4.33.1)

diversion-ratiologit-demandp2o-batch-pfp2ap2o-gran-per-chapterp2o-plan-paperp2o-v1

Consider JJJ platforms with Logit demand nk(t)=e−tk/(w+∑ie−ti)n^k(t) = e^{-t^k}/(w + \sum_{i} e^{-t^i})nk(t)=e−tk/(w+∑i​e−ti), outside-option weight w=ez≥0w = e^{z} \ge 0w=ez≥0, and assume that w>0w > 0w>0 or J≥2J \ge 2J≥2. Fix a platform jjj and a net-fee profile ttt. Then the diversion ratio φj(t)=∑k≠jnk(t)/(1−nj(t))\varphi^j(t) = \sum_{k \ne j} n^k(t)/(1 - n^j(t))φj(t)=∑k=j​nk(t)/(1−nj(t)) satisfies:

  1. the closed form
φj(t)=∑k≠je−tkw+∑k≠je−tk,\varphi^j(t) = \frac{\sum_{k\ne j} e^{-t^k}}{w + \sum_{k \ne j} e^{-t^k}},φj(t)=w+∑k=j​e−tk∑k=j​e−tk​,

which does not involve tjt^jtj; 2. the defining property of the diversion ratio (p. 25): as a function of platform jjj's own net fee sss, with the other fees fixed, the total rival demand ∑k≠jnk\sum_{k\ne j} n^k∑k=j​nk has derivative at s=tjs = t^js=tj

∑k≠j∂nk∂tj=φj(t) (−∂nj∂tj)=φj(t) nj(t)(1−nj(t)).\sum_{k\ne j} \frac{\partial n^k}{\partial t^j} = \varphi^j(t)\,\Big(-\frac{\partial n^j}{\partial t^j}\Big) = \varphi^j(t)\, n^j(t)\big(1 - n^j(t)\big).k=j∑​∂tj∂nk​=φj(t)(−∂tj∂nj​)=φj(t)nj(t)(1−nj(t)).

Independence from tjt^jtj is what makes the condition ∂φj/∂tj≤0\partial\varphi^j/\partial t^j \le 0∂φj/∂tj≤0 of Appendix E.1.3 automatic under Logit demand, so that the first-order condition characterizes best responses.

Formalization Note The Logit own-fee derivative −∂nj/∂tj=nj(1−nj)-\partial n^j/\partial t^j = n^j(1-n^j)−∂nj/∂tj=nj(1−nj) is used to write the right-hand side of part 2. The hypothesis "w>0w > 0w>0 or J≥2J \ge 2J≥2" rules out the degenerate single platform without outside option, where 1−nj=01 - n^j = 01−nj=0.

Preamble
import Mathlib
import Definitions.Def_NetFeePlatform_Competition_Setting
Formal statement
namespace NetFeePlatform.InteropFee
theorem diversion_logit {J : ℕ} (w : ℝ) (hw : 0 ≤ w) (hJ : 0 < w ∨ 2 ≤ J)
    (t : Fin J → ℝ) (j : Fin J) :
    NetFeePlatform.Competition.diversion w t j = (∑ k ∈ Finset.univ.erase j, Real.exp (-t k)) /
        (w + ∑ k ∈ Finset.univ.erase j, Real.exp (-t k)) ∧
      HasDerivAt (fun s => ∑ k ∈ Finset.univ.erase j, NetFeePlatform.Competition.demand w (Function.update t j s) k)
        (NetFeePlatform.Competition.diversion w t j * (NetFeePlatform.Competition.demand w t j * (1 - NetFeePlatform.Competition.demand w t j))) (t j) := by sorry
end NetFeePlatform.InteropFee
Source
Ekmekci, White & Wu, Platform Competition and Interoperability: The Net Fee Model, accepted working-paper version (December 12, 2024), p. 39, Appendix C.2, (32); p. 25, definition of the diversion ratio φʲ(t)
Human review
  • Endorsed by Shuze Chen · Oct 14, 2026

    Confirmed by the moderator at approval.

  • Endorsed by mikedeng1 · Oct 14, 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