Regularity derived from the classical copula conditions #
Rectangle increments split additively along each coordinate. Positivity then gives monotonicity and the Lipschitz estimate; continuity is a consequence of the classical conditions, rather than an extra hypothesis.
theorem
ProbabilityTheory.Copula.partialIncrement_congr_lower
{d : ℕ}
{F : (Fin d → ↑unitInterval) → ℝ}
(a a' b : Fin d → ↑unitInterval)
(s : Finset (Fin d))
(ha : ∀ i ∈ s, a i = a' i)
:
theorem
ProbabilityTheory.Copula.rectangleIncrement_split
{d : ℕ}
{F : (Fin d → ↑unitInterval) → ℝ}
(a b : Fin d → ↑unitInterval)
(i : Fin d)
(t : ↑unitInterval)
:
rectangleIncrement F a b = rectangleIncrement F a (Function.update b i t) + rectangleIncrement F (Function.update a i t) b
Finite additivity under a cut, including cuts at the endpoints.
theorem
ProbabilityTheory.Copula.rectangleIncrement_eq_zero_of_eq
{d : ℕ}
{F : (Fin d → ↑unitInterval) → ℝ}
(a b : Fin d → ↑unitInterval)
(i : Fin d)
(hi : a i = b i)
:
theorem
ProbabilityTheory.Copula.IsClassical.partialIncrement_zero_lower
{d : ℕ}
{F : (Fin d → ↑unitInterval) → ℝ}
(hF : IsClassical F)
(a b : Fin d → ↑unitInterval)
(s : Finset (Fin d))
(ha : ∀ i ∈ s, a i = 0)
:
theorem
ProbabilityTheory.Copula.IsClassical.rectangleIncrement_zero_lower
{d : ℕ}
{F : (Fin d → ↑unitInterval) → ℝ}
(hF : IsClassical F)
(b : Fin d → ↑unitInterval)
:
theorem
ProbabilityTheory.Copula.IsClassical.rectangleIncrement_slab
{d : ℕ}
{F : (Fin d → ↑unitInterval) → ℝ}
(hF : IsClassical F)
(b : Fin d → ↑unitInterval)
(i : Fin d)
(t : ↑unitInterval)
:
rectangleIncrement F (Function.update (fun (x : Fin d) => 0) i t) b = F b - F (Function.update b i t)
theorem
ProbabilityTheory.Copula.IsClassical.rectangleIncrement_mono_upper
{d : ℕ}
{F : (Fin d → ↑unitInterval) → ℝ}
(hF : IsClassical F)
(a b c : Fin d → ↑unitInterval)
(hab : a ≤ b)
(hbc : b ≤ c)
:
Rectangle mass increases when its upper endpoint increases.
theorem
ProbabilityTheory.Copula.IsClassical.monotone
{d : ℕ}
{F : (Fin d → ↑unitInterval) → ℝ}
(hF : IsClassical F)
:
Monotone F
Coordinatewise monotonicity follows from the rectangle condition.
theorem
ProbabilityTheory.Copula.IsClassical.update_sub_le
{d : ℕ}
{F : (Fin d → ↑unitInterval) → ℝ}
(hF : IsClassical F)
(u : Fin d → ↑unitInterval)
(i : Fin d)
(t : ↑unitInterval)
(ht : u i ≤ t)
:
theorem
ProbabilityTheory.Copula.IsClassical.sub_le_sum_of_le
{d : ℕ}
{F : (Fin d → ↑unitInterval) → ℝ}
(hF : IsClassical F)
(a b : Fin d → ↑unitInterval)
(hab : a ≤ b)
:
theorem
ProbabilityTheory.Copula.IsClassical.sub_le_sum_abs
{d : ℕ}
{F : (Fin d → ↑unitInterval) → ℝ}
(hF : IsClassical F)
(a b : Fin d → ↑unitInterval)
:
theorem
ProbabilityTheory.Copula.IsClassical.abs_sub_le_sum_abs
{d : ℕ}
{F : (Fin d → ↑unitInterval) → ℝ}
(hF : IsClassical F)
(a b : Fin d → ↑unitInterval)
:
The sharp Lipschitz estimate in the sum of coordinate distances.
theorem
ProbabilityTheory.Copula.IsClassical.lipschitzWith
{d : ℕ}
{F : (Fin d → ↑unitInterval) → ℝ}
(hF : IsClassical F)
:
LipschitzWith (↑d) F
theorem
ProbabilityTheory.Copula.IsClassical.continuous
{d : ℕ}
{F : (Fin d → ↑unitInterval) → ℝ}
(hF : IsClassical F)
: