Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Theorem 1, p. 12 — optimal prices equalize aggregate markups; the optimal profit is the unique fixed point of H(z) = sup_r H(r; z), strictly decreasing for S ≠ ∅

Proved
CrossSell.Pricing.theorem_1

by mikedeng1 · Oct 22, 2026 · Mathlib 0df444a (Lean v4.33.1)

aggregate-markupfixed-pointmultinomial-logitp2o-batch-pfp2ap2o-gran-per-chapterp2o-plan-paperp2o-v1pricing

Consider the sequential multinomial logit model of cross-category retailing: a firm offers a finite set SSS of main products, with S+=S∪{0}S^+ = S \cup \{0\}S+=S∪{0}; after the first-stage outcome i∈S+i \in S^+i∈S+ a consumer may buy one secondary product from TiT_iTi​, chosen from finitely many secondary products. Main product iii has intrinsic utility uiu_iui​ and cost cic_ici​; secondary product lll has cost sls_lsl​ and intrinsic utility vilv_{il}vil​ after outcome iii. With main prices p\mathbf pp and secondary prices r\mathbf rr the total expected profit is Π(S,T,p,r)\Pi(S,\mathbf T,\mathbf p,\mathbf r)Π(S,T,p,r) of eq. (5), the secondary profit of TiT_iTi​ is

ti(r)=∑l∈Ti(rl−sl)exp⁡(vil−rl)1+∑k∈Tiexp⁡(vik−rk),t_i(\mathbf r) = \sum_{l\in T_i}\frac{(r_l - s_l)\exp(v_{il} - r_l)}{1 + \sum_{k\in T_i}\exp(v_{ik} - r_k)},ti​(r)=l∈Ti​∑​1+∑k∈Ti​​exp(vik​−rk​)(rl​−sl​)exp(vil​−rl​)​,

and

H(r;z)=∑i∈Sexp⁡((ui−ci−1−z)+ti(r))+t0(r),H(z)=sup⁡rH(r;z).H(\mathbf r; z) = \sum_{i\in S}\exp\big((u_i - c_i - 1 - z) + t_i(\mathbf r)\big) + t_0(\mathbf r), \qquad H(z) = \sup_{\mathbf r} H(\mathbf r; z).H(r;z)=i∈S∑​exp((ui​−ci​−1−z)+ti​(r))+t0​(r),H(z)=rsup​H(r;z).

Then:

  1. (a) at every joint maximizer (p,r)(\mathbf p, \mathbf r)(p,r) of Π\PiΠ, the aggregate markup pi−ci+ti(r)p_i - c_i + t_i(\mathbf r)pi​−ci​+ti​(r) is the same for all i∈Si \in Si∈S;
  2. for every zzz, {H(r;z):r}\{H(\mathbf r; z) : \mathbf r\}{H(r;z):r} is bounded above, so H(z)H(z)H(z) is finite;
  3. (b) the equation z=H(z)z = H(z)z=H(z) has exactly one solution z∗z^*z∗, and z∗z^*z∗ is the optimal total profit: the least upper bound of {Π(S,T,p,r):p,r}\{\Pi(S,\mathbf T,\mathbf p,\mathbf r) : \mathbf p, \mathbf r\}{Π(S,T,p,r):p,r};
  4. if S≠∅S \neq \emptysetS=∅, H(z)H(z)H(z) is strictly decreasing in zzz.

The theorem reduces the joint pricing problem, which has up to M+NM + NM+N variables and is not concave, to a secondary-price problem max⁡rH(r;z)\max_{\mathbf r} H(\mathbf r; z)maxr​H(r;z) and a one-dimensional fixed-point search.

Formalization Note. "Optimality" is global maximization of Π\PiΠ over all real price vectors with SSS and T\mathbf TT fixed. The optimal total profit is stated as the least upper bound of the profit values (the paper's "max" presumes attainment, which the page does not show; the companion milestone on the p. 13 price formula gives attainment whenever H(⋅;z∗)H(\cdot; z^*)H(⋅;z∗) has a maximizer). H(z)H(z)H(z) is a real supremum, made meaningful by clause 2. Clause 4 carries the hypothesis S≠∅S \neq \emptysetS=∅, not printed on the page: for S=∅S = \emptysetS=∅, HHH is constant, so strict decrease fails; the other clauses hold for every SSS.

Preamble
import Mathlib
import Definitions.Def_CrossSell_Pricing_Model
Formal statement
namespace CrossSell.Pricing

theorem theorem_1 {ι κ : Type*} [Fintype κ] (S : Finset ι) (T : Option ι → Finset κ)
    (u c : ι → ℝ) (s : κ → ℝ) (v : Option ι → κ → ℝ) :
    (∀ (p : ι → ℝ) (r : κ → ℝ),
        (∀ (p' : ι → ℝ) (r' : κ → ℝ), profit S T u c s v p' r' ≤ profit S T u c s v p r) →
        ∀ i ∈ S, ∀ j ∈ S, aggMarkup T c s v p r i = aggMarkup T c s v p r j) ∧
    (∀ z : ℝ, BddAbove (Set.range fun r : κ → ℝ => Hr S T u c s v r z)) ∧
    (∃ zstar : ℝ, zstar = Hmax S T u c s v zstar ∧
        (∀ z : ℝ, z = Hmax S T u c s v z → z = zstar) ∧
        IsLUB (Set.range fun pr : (ι → ℝ) × (κ → ℝ) => profit S T u c s v pr.1 pr.2) zstar) ∧
    (S.Nonempty → StrictAnti (Hmax S T u c s v)) := by sorry

end CrossSell.Pricing
Source
Ke & Wang, Cross-Category Retailing Management: Substitution and Complementarity, accepted manuscript (SSRN 3383023), p. 12, Theorem 1
Human review
  • Endorsed by Shuze Chen · Oct 23, 2026

    Confirmed by the moderator at approval.

  • Endorsed by mikedeng1 · Oct 23, 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