Lemma EC.4 — ordered-weights mediant inequality
ProvedStaticCalendar.lemma_EC_4assortment-optimizationdynamic-pricingp2o-gran-per-chapterp2o-plan-paperp2o-v1revenue-management
Let for with , and let . Then
Weights that decrease together with the ratios can only raise the mediant. It is the tool behind (EC.5).
Differences from the source:
- encoding: no condition that some is positive → hypothesis ; reason: it excludes only the case in which the left side is
Preamble
import Mathlib
Formal statement
namespace StaticCalendar
/-- Lemma EC.4, arXiv:1811.01077v5, p.ec4: for `aᵢ, bᵢ > 0` with `a₁/b₁ ≥ ⋯ ≥ aₙ/bₙ` and
`β₁ ≥ ⋯ ≥ βₙ ≥ 0`, `Σ βᵢaᵢ / Σ βᵢbᵢ ≥ Σ aᵢ / Σ bᵢ`.
**Formalization Note.** The hypothesis `∃ i, 0 < β i` excludes only the case in which the printed
left side is `0/0`. -/
theorem lemma_EC_4 {n : ℕ} (a b β : Fin n → ℝ) (ha : ∀ i, 0 < a i) (hb : ∀ i, 0 < b i)
(hratio : Antitone fun i => a i / b i) (hβ : Antitone β) (hβ0 : ∀ i, 0 ≤ β i)
(hβpos : ∃ i, 0 < β i) :
(∑ i, a i) / (∑ i, b i) ≤ (∑ i, β i * a i) / (∑ i, β i * b i) := by sorry
end StaticCalendar
Source
Ma, Simchi-Levi & Zhao, Dynamic Pricing (and Assortment) under a Static Calendar, arXiv:1811.01077v5, Lemma EC.4, p.ec4
Human review
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.