Eq. (39a) — feasible inverse substitution when t is positive
ProvedChanceDetEquiv.PModel.backward_39achance-constrained-programmingfractional-programmingp2o-batch-pfp2bp2o-gran-per-chapterp2o-plan-paperp2o-v1perspective-transform
For positive row and decision dimensions, let satisfy (39), and suppose . Divide the matrix, row slacks, and objective numerator by and put . The resulting satisfies the corrected program (38), with
This identifies the positive- part of (39) with the fractional program. The points at remain in (39) and are handled by the supremum theorem.
Formalization Note The condition is essential to division. The moments are the raw expected squares of (30), (34), and (39b), and . Each row residual is assumed Gaussian, as in the source.
Preamble
import Definitions.Def_ChanceDetEquiv_PModel_Model
Formal statement
namespace ChanceDetEquiv.PModel
open MeasureTheory
/-- For t > 0, the substitution (39a) has a feasible inverse preserving objective values. -/
theorem backward_39a {Ω : Type*} [MeasurableSpace Ω] {m n : ℕ}
(d : Model Ω m n) [IsProbabilityMeasure d.P]
(hm : 0 < m) (hn : 0 < n)
(hMoments : MomentAssumptions d) (hNormal : NormalRows d)
(hConfidence : Confidence d)
(p : Point39 m n) (hp : p ∈ feasible39 d) (ht : 0 < p.2.2.2.2) :
backward p ∈ feasible38 d ∧ value38 (backward p) = value39 p := by sorry
end ChanceDetEquiv.PModel
Source
Charnes and Cooper, Deterministic Equivalents for Optimizing and Satisficing under Chance Constraints, Oper. Res. 11 (1963), p. 32, Eq. (39a), read in reverse; https://doi.org/10.1287/opre.11.1.18
Human review
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.