Realizing a law of conditional laws by a copula #
noncomputable def
Verification.copulaOfKernel
(κ : ProbabilityTheory.Kernel ↑unitInterval ↑unitInterval)
[ProbabilityTheory.IsMarkovKernel κ]
(hκ : MeasureTheory.volume.bind ⇑κ = MeasureTheory.volume)
:
A Markov kernel preserving the uniform law defines a copula.
Equations
- One or more equations did not get rendered due to their size.
Instances For
theorem
Verification.copulaOfKernel_conditional
(κ : ProbabilityTheory.Kernel ↑unitInterval ↑unitInterval)
[ProbabilityTheory.IsMarkovKernel κ]
(hκ : MeasureTheory.volume.bind ⇑κ = MeasureTheory.volume)
:
⇑(copulaOfKernel κ hκ).conditionalKernel =ᵐ[MeasureTheory.volume] ⇑κ
Its selected conditional kernel is the original kernel almost everywhere.
theorem
Verification.exists_copula_conditionalLaw
(Λ : MeasureTheory.ProbabilityMeasure (MeasureTheory.ProbabilityMeasure ↑unitInterval))
(hΛ : Λ ∈ uniformBarycenter)
:
∃ (C : ProbabilityTheory.Copula 2), conditionalLaw C = Λ
Every law of probability measures with uniform barycenter is realized by a copula.