Proof of Proposition 3.15, p. 17 — some γ ∈ (0, 1] gives D(τ₁ − τ₂) = 0 for τᵢ := γµᵢ + (1 − γ)δ(xᵢ)
ProvedDistCov.Converse.exists_gamma_Ddiff_eq_zerometric-spacenegative-typep2o-batch-pfp2ap2o-gran-per-chapterp2o-plan-paperp2o-v1
Let be a separable metric space, let be Borel probability measures on with finite first moments and , and let be points of . For put , . Then there is such that
This is the step that produces, on a space of negative type which is not of strong negative type (and also off negative type), a pair of probability measures whose energy difference vanishes, to be fed into the two-point law of the proof of Proposition 3.15.
Formalization Note. Separability is the standing assumption of Errata (i). The weights , multiply measures as extended nonnegative reals; this is faithful because .
Preamble
import Mathlib import Definitions.Def_DistCov_Converse_Setting open MeasureTheory
Formal statement
namespace DistCov.Converse
theorem exists_gamma_Ddiff_eq_zero
{X : Type*} [MetricSpace X] [MeasurableSpace X] [BorelSpace X] [SecondCountableTopology X]
(μ₁ μ₂ : Measure X) [IsProbabilityMeasure μ₁] [IsProbabilityMeasure μ₂]
(hμ₁ : DistCov.Indep.FiniteFirstMoment μ₁) (hμ₂ : DistCov.Indep.FiniteFirstMoment μ₂)
(hD : 0 ≤ DistCov.Indep.Ddiff μ₁ μ₂) (x₁ x₂ : X) (hx : x₁ ≠ x₂) :
∃ γ ∈ Set.Ioc (0 : ℝ) 1, DistCov.Indep.Ddiff (mixDirac γ μ₁ x₁) (mixDirac γ μ₂ x₂) = 0 := by sorry
end DistCov.Converse
Source
Lyons, Distance covariance in metric spaces, arXiv:1106.5758 (version of 19 Dec. 2020), p. 17, proof of Proposition 3.15, second paragraph
Human review
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.