Theorem 2.3, (6), p. 8, case d = 0 — ∅ ≠ S_U(ν) ⊆ S(μ) + Ψ(L̂ d_{F_U}(μ, ν))𝔹 for each ν ∈ P_{F_U}
ProvedProbMetricStab.General.theorem_2_3_d_zerop2o-batch-pfp2ap2o-gran-per-chapterp2o-plan-paperp2o-v1quantitative-stabilitysolution-setsstochastic-programming
Consider model (1) without stochastic constraints () and assume the hypotheses of Theorem 2.2 (the general assumptions, , , an open bounded neighbourhood of ). Let be the growth function of problem (1) on ,
and . Then there is a constant such that for each
This is Theorem 2.3 exactly as printed, in the case , where it holds for all perturbations without any smallness condition: the solution sets move by at most of the minimal information distance.
Formalization Note , and take values in ; when the inclusion is trivial. The paper's "min" in is read as an infimum ( over the empty set). For the printed "for each " fails; see the goal theorem theorem_2_3.
Preamble
import Mathlib import Definitions.Def_ProbMetricStab_General_Setting open MeasureTheory Set Metric Filter Topology open scoped ENNReal
Formal statement
namespace ProbMetricStab.General
theorem theorem_2_3_d_zero {m s : ℕ} {P : Model m s 0}
{U : Set (EuclideanSpace ℝ (Fin m))} {μ : Measure ↥P.Ξ} (h : P.Thm22Assumptions U μ) :
∃ Lhat : ℝ, 1 ≤ Lhat ∧ ∀ ν ∈ P.PFU U,
(P.SU U ν).Nonempty ∧
P.SU U ν ⊆ enlarge (P.S μ) (P.Ψ U μ (ENNReal.ofReal Lhat * P.dFU U μ ν)) := by sorry
end ProbMetricStab.General
Source
Rachev & Römisch, Quantitative stability in stochastic programming: The method of probability metrics, preprint (edoc.hu-berlin.de), p. 8, Theorem 2.3, (6), and its proof ("In case that d = 0 ...")
Human review
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.