Tables 1–2: Nelsen 16 constructor and zero endpoint #
theorem
Papers.AnsariRockel2024.nelsen16_cdf_continuous_parameter
(u v : ↑unitInterval)
:
Continuous fun (θ : ↑(Set.Ici 0)) => (Verification.nelsen16 ↑θ ⋯).cdf ![u, v]
theorem
Papers.AnsariRockel2024.nelsen16_tendsto_zero
{α : Type u_1}
{l : Filter α}
(θ : α → ℝ)
(hθ : ∀ (a : α), 0 ≤ θ a)
(ht : Filter.Tendsto θ l (nhds 0))
(u v : ↑unitInterval)
:
Filter.Tendsto (fun (a : α) => (Verification.nelsen16 (θ a) ⋯).cdf ![u, v]) l
(nhds (ProbabilityTheory.Copula.countermonotonic.cdf ![u, v]))
theorem
Papers.AnsariRockel2024.nelsen16_lowerOrthant_monotone
{θ η : ℝ}
(hθ : 0 ≤ θ)
(hη : 0 ≤ η)
(hθη : θ ≤ η)
:
(Verification.nelsen16 θ hθ).LowerOrthantLE (Verification.nelsen16 η hη)
theorem
Papers.AnsariRockel2024.nelsen16_tendsto_atTop
{α : Type u_1}
{l : Filter α}
(θ : α → ℝ)
(hθ : ∀ (z : α), 0 ≤ θ z)
(hlim : Filter.Tendsto θ l Filter.atTop)
(u v : ↑unitInterval)
:
Filter.Tendsto (fun (z : α) => (Verification.nelsen16 (θ z) ⋯).cdf ![u, v]) l
(nhds ((ProbabilityTheory.Copula.clayton 2 1 ⋯).cdf ![u, v]))
theorem
Papers.AnsariRockel2024.nelsen16_tails
(θ : ℝ)
(hθ : 0 ≤ θ)
:
(Verification.nelsen16 θ hθ).HasLowerTailDependence (if θ = 0 then 0 else 1 / 2) ∧ (Verification.nelsen16 θ hθ).HasUpperTailDependence 0
theorem
Papers.AnsariRockel2024.nelsen16_schur_monotone
{θ η : ℝ}
(hθ : 3 ≤ θ)
(hη : 3 ≤ η)
(hθη : θ ≤ η)
:
(Verification.nelsen16 θ ⋯).SchurBothLE (Verification.nelsen16 η ⋯)
theorem
Papers.AnsariRockel2024.nelsen16_toMeasure_density
{θ : ℝ}
(hθ : 0 < θ)
:
(Verification.nelsen16 θ ⋯).toMeasure = MeasureTheory.volume.withDensity fun (x : Fin 2 → ↑unitInterval) => ENNReal.ofReal (Verification.n16Density θ x)