Proof of Proposition 7, PDF p. 47 — the SQR-heter revenue is at least the first-stage revenue
ProvedMultiStoreAssort.Approx.first_inequalityIn the heterogeneous multi-store model ( stores with arrival fractions , ; products with revenues ; attraction values of product in store ; utility discounts with ), let be the (SQR-heter) revenue of a profile , with the sequential choice probabilities (10). Then for every profile ,
The right-hand side is the revenue of the first stage, in which each customer considers only the assortment of the store she visits. In the proof of Proposition 7 this is applied to the heuristic profile , where it reads .
Formalization Note. The paper writes the inequality for ; it is stated here for every profile, a generalization the paper's argument covers unchanged. The positivity of the revenues is a pinned hypothesis: the inequality drops the second-stage revenue, which is nonnegative because revenues are prices; the paper never states the sign.
import Mathlib import Definitions.Def_ChoiceCDLP_MNL_mnlObjective import Definitions.Def_MultiStoreAssort_Approx_Model
namespace MultiStoreAssort.Approx
/-- First inequality of the proof of Proposition 7 (Online Supplement p. 15), stated for every
profile `T`: the (SQR-heter) revenue of `T` is at least the revenue of its first stage,
`∑_i λ_i ∑_{k ∈ T_i} r_k v_{ik} / (1 + ∑_{k ∈ T_i} v_{ik})`. -/
theorem first_inequality {m n : ℕ} (lam : Fin m → ℝ) (r : Fin n → ℝ) (v : Fin m → Fin n → ℝ)
(c : Fin m → Fin m → Fin n → ℝ)
(hlam : ∀ i, 0 ≤ lam i) (hlam1 : ∑ i, lam i = 1) (hr : StrictAnti r) (hrpos : ∀ k, 0 < r k)
(hv : ∀ i k, 0 < v i k) (hc : ∀ i j k, 0 ≤ c i j k) (hcii : ∀ i k, c i i k = 0) :
∀ T : Fin m → Finset (Fin n),
∑ i, lam i * MultiStoreAssort.SMR.single r (v i) (T i) ≤ revQh lam r v c T := by sorry
end MultiStoreAssort.Approx
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.