Documentation

Papers.OrendayLaresRockel2026XiBeta.RBranchLaw

← Mathematical handbook

The law of (U, V_b) (Proposition 5.3, first statement) #

U ~ U(0,1) and an independent fair coin ε (Bool); V_b = U if U ∉ (b/2, 1 - b/2) and V_b = (2U + b)/4 + ε (1 - b)/2 otherwise. We show that the law of (U, V_b) is the measure of the copula R_b, so that R_b is the copula of (U, V_b).

noncomputable def Papers.OrendayLaresRockel2026XiBeta.vbReal (b u : ℝ) (e : Bool) :

The real-valued formula for V_b as a function of (U, ε).

Equations
Instances For
    theorem Papers.OrendayLaresRockel2026XiBeta.vbReal_mem {b : ℝ} (hb : b ∈ Set.Icc 0 1) {u : ℝ} (hu : u ∈ Set.Icc 0 1) (e : Bool) :
    vbReal b u e ∈ Set.Icc 0 1
    theorem Papers.OrendayLaresRockel2026XiBeta.coe_Vb {b : ℝ} (hb : b ∈ Set.Icc 0 1) (u : ↑unitInterval) (e : Bool) :
    ↑(Vb b u e) = vbReal b (↑u) e

    The random vector (U, V_b) on I × Bool.

    Equations
    Instances For
      theorem Papers.OrendayLaresRockel2026XiBeta.rightSection_B {b : ℝ} (hb : b ∈ Set.Icc 0 1) {t : ℝ} (ht : 0 ≤ t) :
      (rightSection b hb).B t = 1 / 2 + t - rbA b t
      theorem Papers.OrendayLaresRockel2026XiBeta.kerR_rightSection {b : ℝ} (hb : b ∈ Set.Icc 0 1) {t v : ℝ} (ht : t ∈ Set.Icc 0 1) (hne : t ≠ 1 - b / 2) :
      (rightSection b hb).kerR v t = ((if vbReal b t false ≤ v then 1 else 0) + if vbReal b t true ≤ v then 1 else 0) / 2

      The kernel of R_b at t ≠ 1 - b/2 is the law of V_b given U = t.

      theorem Papers.OrendayLaresRockel2026XiBeta.cdf_lawVb {b : ℝ} (hb : b ∈ Set.Icc 0 1) (u v : ↑unitInterval) :
      (lawVb b).real (Set.Iic ![u, v]) = (Rb b hb).cdf ![u, v]

      The distribution function of the law of (U, V_b) is that of R_b.

      Two probability measures on the unit square with equal distribution functions coincide.

      Proposition 5.3, first statement: the measure of R_b is the law of (U, V_b).