Proof of Theorem 5.2, p. 25 — |r_{α,Γ}(x) − r_{α̃,Γ̃}(x)| ≤ αζ₁(Γ, Γ̃) + e⁻¹|α − α̃| on X
ProvedProbMetricStab.Portfolio.risk_lipschitzp2o-batch-pfp2ap2o-gran-per-chapterp2o-plan-paperp2o-v1probability-metricsquantitative-stabilitystochastic-programming
Let and let be Borel probability measures on the unit sphere . With , the standard simplex and the dual Lipschitz metric on , for every
The risk is thus jointly Lipschitz in the stability index and the spectral measure, uniformly on ; together with the growth condition of Lemma 5.1 this is what Theorem 5.2 needs.
Formalization Note Both sides are compared in because is defined as a supremum in ; it is finite (at most ), so the inequality is the real one.
Preamble
import Mathlib import Definitions.Def_ProbMetricStab_Portfolio_Setting open MeasureTheory open scoped ENNReal
Formal statement
namespace ProbMetricStab.Portfolio
theorem risk_lipschitz {s : ℕ} (α α' : ℝ) (hα1 : 1 < α) (hα2 : α < 2) (hα'1 : 1 < α')
(hα'2 : α' < 2) (Γ Γ' : Measure (Rs s)) (hΓ : IsSpectral Γ) (hΓ' : IsSpectral Γ')
(x : Rs s) (hx : x ∈ simplex s) :
ENNReal.ofReal |risk α Γ x - risk α' Γ' x|
≤ ENNReal.ofReal α * zeta1 Γ Γ' + ENNReal.ofReal (Real.exp (-1) * |α - α'|) := by sorry
end ProbMetricStab.Portfolio
Source
Rachev & Römisch, Quantitative stability in stochastic programming: The method of probability metrics, preprint (edoc.hu-berlin.de), p. 25, proof of Theorem 5.2, first display (outer inequality)
Human review
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.