Proposition 4.1(ii), p. 934 — for α, β ≥ 0, αg₁ + βg₂ is (αK + βL)-convex
ProvedSethiChengSS.Finite.proposition_4_1_iik-convexityp2o-batch-pfp2ap2o-gran-per-chapterp2o-plan-paperp2o-v1
Let , let be -convex and be -convex. Then for all ,
In the proof of Theorem 4.1 this carries -convexity through the weighted sum in (3.1) and through the addition of the linear term .
Formalization Note The published BertsekasDP.kconvex_combination requires . The paper allows zero coefficients, which matter here because transition probabilities may vanish.
Preamble
import Mathlib import Definitions.Def_BertsekasKConvex import Definitions.Def_SethiChengSS_Finite_KConvexity open MeasureTheory Filter Topology
Formal statement
namespace SethiChengSS.Finite
/-- Proposition 4.1(ii) (Sethi–Cheng 1997, p. 934): if `g₁` is `K`-convex and `g₂` is
`L`-convex, then for `α, β ≥ 0` the function `α g₁ + β g₂` is `(αK + βL)`-convex. -/
theorem proposition_4_1_ii (K L α β : ℝ) (g₁ g₂ : ℝ → ℝ) (hK : 0 ≤ K) (hL : 0 ≤ L)
(hα : 0 ≤ α) (hβ : 0 ≤ β) (h₁ : BertsekasKConvex K g₁) (h₂ : BertsekasKConvex L g₂) :
BertsekasKConvex (α * K + β * L) (fun x => α * g₁ x + β * g₂ x) := by sorry
end SethiChengSS.Finite
Source
Sethi and Cheng, Optimality of (s, S) policies in inventory models with Markovian demand, Oper. Res. 45(6) (1997), DOI 10.1287/opre.45.6.931, p. 934, Proposition 4.1(ii)
Human review
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.