Spearman rho dominates xi under stochastic monotonicity #
The scalar estimate compares squared differences with absolute differences of an antitone function taking values in [0,1].
theorem
ProbabilityTheory.Copula.RankRegion.XiRho.Support.integral_lower_integral
{g : ↑unitInterval → ℝ}
(hg : Measurable g)
(hb : ∀ (u : ↑unitInterval), g u ∈ Set.Icc 0 1)
:
∫ (u : ↑unitInterval), ∫ (w : ↑unitInterval) in Set.Iic u, g w = ∫ (w : ↑unitInterval), (1 - ↑w) * g w
theorem
ProbabilityTheory.Copula.RankRegion.XiRho.Support.integrable_lower_integral
{g : ↑unitInterval → ℝ}
(hg : Measurable g)
(hb : ∀ (u : ↑unitInterval), g u ∈ Set.Icc 0 1)
:
MeasureTheory.Integrable (fun (u : ↑unitInterval) => ∫ (w : ↑unitInterval) in Set.Iic u, g w) MeasureTheory.volume
theorem
ProbabilityTheory.Copula.RankRegion.XiRho.Support.integral_abs_sub_of_antitone
{g : ↑unitInterval → ℝ}
(hg : Antitone g)
(hb : ∀ (u : ↑unitInterval), g u ∈ Set.Icc 0 1)
(u : ↑unitInterval)
:
theorem
ProbabilityTheory.Copula.RankRegion.XiRho.Support.integral_sq_le_twice_lower_integral
{g : ↑unitInterval → ℝ}
(hg : Antitone g)
(hb : ∀ (u : ↑unitInterval), g u ∈ Set.Icc 0 1)
:
∫ (u : ↑unitInterval), g u ^ 2 ≤ ((2 * ∫ (u : ↑unitInterval), ∫ (w : ↑unitInterval) in Set.Iic u, g w) - ∫ (u : ↑unitInterval), g u) + (∫ (u : ↑unitInterval), g u) ^ 2
theorem
ProbabilityTheory.Copula.RankRegion.XiRho.Support.conditionalCDF_sq_le_cdf_integral
(C : Copula 2)
(hC : C.IsSI)
(v : ↑unitInterval)
:
theorem
ProbabilityTheory.Copula.RankRegion.XiRho.Support.xi_le_rho_of_isSI
(C : Copula 2)
(hC : C.IsSI)
:
theorem
ProbabilityTheory.Copula.RankRegion.XiRho.Support.xi_le_neg_rho_of_isSD
(C : Copula 2)
(hC : C.IsSD)
:
theorem
ProbabilityTheory.Copula.RankRegion.XiRho.Support.xi_le_abs_rho_of_stochastically_monotone
(C : Copula 2)
(hC : C.IsSI ∨ C.IsSD)
: