Proposition 7.1 — positive homogeneity with subadditivity is equivalent to positive homogeneity with convexity
ProvedKallMayerSLP.RiskMeasures.proposition_7_1convex-analysisp2o-gran-per-chapterp2o-plan-bookp2o-v1risk-measuresstochastic-programming
Let be a function on a real linear space . Then:
- if is positive homogeneous and subadditive, then is convex;
- if is positive homogeneous and convex, then is subadditive.
Here positive homogeneity means for all , , and subadditivity for all . The proposition is the bridge between the axiom systems (7.2) and (7.3): a convex risk measure is coherent exactly when it is positively homogeneous.
Differences from the source: none.
Preamble
import Mathlib import Definitions.Def_KallMayerSLP_RiskMeasures_IsPosHomogeneous import Definitions.Def_KallMayerSLP_RiskMeasures_IsSubadditive open MeasureTheory open scoped ENNReal
Formal statement
namespace KallMayerSLP.RiskMeasures
/-- **Proposition 7.1** (sublinear functions).
Kall & Mayer, *Stochastic Linear Programming* (2005), Chapter 2, §7, Prop. 7.1, p.181 (PDF 191). Let `g : X → ℝ` be a function on a linear space `X`. Then
a) if `g` is both positive homogeneous and subadditive then it is convex;
b) if `g` is positive homogeneous and convex, then it is subadditive.
**Formalization Note.** `X` is any real vector space; convexity is `ConvexOn ℝ univ g`. -/
theorem proposition_7_1 {X : Type*} [AddCommGroup X] [Module ℝ X] (g : X → ℝ) :
(IsPosHomogeneous g → IsSubadditive g → ConvexOn ℝ Set.univ g) ∧
(IsPosHomogeneous g → ConvexOn ℝ Set.univ g → IsSubadditive g) := by sorry
end KallMayerSLP.RiskMeasures
Source
Kall & Mayer, Stochastic Linear Programming (2005), Chapter 2, §7, Proposition 7.1, p.181 (PDF 191)
Human review
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.