Proposition 3 —
ProvedProductFraming.Nest.proposition_3bound-revealing-programp2o-batch-pfp2bp2o-gran-per-chapterp2o-plan-paperp2o-v1product-framing
Fix . Let be the optimal value of program (5), and let be independent of with the same law. Then the minimum of (5) is attained, and is the optimal value of program (6):
subject to and for , where and .
This reduces the min–max program (5) to a maximization over NBUE distributions only, which is then bounded by the exponential case.
Formalization Note The statement asserts the existence of that is the least value of the objective of (5) over its feasible set, and that is the greatest value of the objective of (6) over its feasible set (both attained).
Preamble
import Mathlib import Definitions.Def_ProductFraming_Nest_Program
Formal statement
namespace ProductFraming.Nest
open Finset
/-- Proposition 3 (Gallego, Li, Truong, Wang 2020, p. 10): the minimum `γ` of (5) is attained, and
`1/γ` is the maximum of `E[X / E[min(X, Y) | X]]` over the tails `Λ` feasible for (6). -/
theorem proposition_3 (m : ℕ) (hm : 1 ≤ m) :
∃ γ : ℝ, IsLeast {J : ℝ | ∃ U Λ : ℕ → ℝ, Feasible5 m U Λ ∧ J = J5 hm U Λ} γ ∧
IsGreatest {t : ℝ | ∃ Λ : ℕ → ℝ, Feasible6 m Λ ∧ t = ratio6 m Λ} (1 / γ) := by sorry
end ProductFraming.Nest
Source
Gallego, Li, Truong, Wang, Approximation Algorithms for Product Framing and Pricing, Operations Research (2020), DOI 10.1287/opre.2019.1875, authors' accepted manuscript, p. 10, Proposition 3 (proof p. 36)
Human review
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.