Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Theorem 4: hm(vj)=πjm(vj)−Lm(vj)h^m(v_j) = \pi_j^m(v_j) - L^m(v_j)hm(vj​)=πjm​(vj​)−Lm(vj​) is quasi-convex in vjv_jvj​ on [0,∞)[0,\infty)[0,∞)

Proved
AssortSearch.MNLQC.theorem_4

by mikedeng1 · Oct 8, 2026 · Mathlib 0df444a (Lean v4.33.1)

assortment-planningmultinomial-logitoperations-researchp2o-batch-pfp2ap2o-gran-per-chapterp2o-plan-paperp2o-v1quasi-convexity

Consider the traditional multinomial logit (MNL) model of consumer choice, without search. There are nnn variants with preferences v1,…,vn>0v_1,\dots,v_n > 0v1​,…,vn​>0 and a no-purchase option with preference v0>0v_0 > 0v0​>0. Variant iii has margin mi∈Rm_i \in \mathbb Rmi​∈R. Including a variant with demand qqq costs c(q)c(q)c(q), where the operational cost ccc is concave and increasing on [0,1][0,1][0,1]. With assortment SSS, variant iii's demand is qim(S)=vi/(∑k∈Svk+v0)q_i^m(S) = v_i/(\sum_{k\in S} v_k + v_0)qim​(S)=vi​/(∑k∈S​vk​+v0​) and its profit is πi(S)=miqi(S)−c(qi(S))\pi_i(S) = m_i q_i(S) - c(q_i(S))πi​(S)=mi​qi​(S)−c(qi​(S)).

Fix an assortment SSS and a variant j∉Sj \notin Sj∈/S, and regard jjj's preference vjv_jvj​ as a variable. Let πi(vj)\pi_i(v_j)πi​(vj​) be variant iii's profit in the assortment S∪{j}S \cup \{j\}S∪{j}, and let L(vj)=∑i∈Sπi(S)−∑i∈Sπi(vj)L(v_j) = \sum_{i\in S}\pi_i(S) - \sum_{i\in S}\pi_i(v_j)L(vj​)=∑i∈S​πi​(S)−∑i∈S​πi​(vj​) be the profit lost on the variants already in SSS. Then the profit change from adding jjj,

hm(vj)=πjm(vj)−Lm(vj),h^m(v_j) = \pi_j^m(v_j) - L^m(v_j),hm(vj​)=πjm​(vj​)−Lm(vj​),

is quasi-convex in vjv_jvj​ on [0,∞)[0,\infty)[0,∞). That is, for every r∈Rr \in \mathbb Rr∈R the set {vj≥0:hm(vj)≤r}\{v_j \ge 0 : h^m(v_j) \le r\}{vj​≥0:hm(vj​)≤r} is an interval.

Because hmh^mhm is quasi-convex, it attains its maximum over any interval [0,vˉ][0, \bar v][0,vˉ] at an endpoint. If adding a less popular variant is profitable, adding a more popular one is even better. This is how the paper shows that the optimal no-search assortment consists of the most popular variants. It extends van Ryzin and Mahajan's (1999) result from their newsvendor cost to any concave increasing cost.

Formalization Note Variants are Fin n, 0-based, and the no-purchase preference v0v_0v0​ is a separate positive real. j∉Sj \notin Sj∈/S is assumed, as the paper's "adding variant jjj to an assortment SSS" requires. ccc is assumed concave and increasing only on [0,1][0,1][0,1], where demands lie, and nothing is assumed about c(0)c(0)c(0), so hm(0)=−c(0)h^m(0) = -c(0)hm(0)=−c(0) need not be 000. No differentiability of ccc is assumed. Margins may differ across variants, as in the paper's model. The paper's proof writes a common margin, but the theorem does not need one.

Preamble
import Mathlib
import Definitions.Def_RetailVariety_Structure_Model
import Definitions.Def_AssortSearch_MNLQC_Model
Formal statement
namespace AssortSearch.MNLQC

/-- **Theorem 4** (p. 13): in the no-search MNL model, with a concave increasing operational
cost `c` on `[0, 1]` and arbitrary margins `m_i`, the function
`h^m(v_j) = π_j^m(v_j) − L^m(v_j)` is quasi-convex in the added variant's preference `v_j`
on `[0, ∞)`. -/
theorem theorem_4
    {n : ℕ} (m : Fin n → ℝ) (c : ℝ → ℝ) (v : Fin n → ℝ) (v0 : ℝ)
    (S : Finset (Fin n)) (j : Fin n)
    (hv : ∀ i, 0 < v i) (hv0 : 0 < v0)
    (hc_concave : ConcaveOn ℝ (Set.Icc 0 1) c) (hc_mono : MonotoneOn c (Set.Icc 0 1))
    (hj : j ∉ S) :
    QuasiconvexOn ℝ (Set.Ici 0) (hm m c v v0 S j) := by sorry

end AssortSearch.MNLQC
Source
Cachon, Terwiesch & Xu, Retail Assortment Planning in the Presence of Consumer Search, working paper (Dec. 20, 2002), p. 13 (PDF 15), Theorem 4
Human review
  • Endorsed by Shuze Chen · Oct 8, 2026

    Confirmed by the moderator at approval.

  • Endorsed by mikedeng1 · Oct 8, 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