Documentation

Atlas.TheoryOfProbability.code.UIEquivalences

def TheoryOfProbability3.TextbookUniformlyIntegrable' {α : Type u_1} {ι : Type u_2} {m : MeasurableSpace α} (f : ια) (μ : MeasureTheory.Measure α) :

Textbook definition of uniform integrability of a family f : ι → α → ℝ: each f i is AEStronglyMeasurable and for every ε > 0 there is a uniform truncation level C : ℝ≥0 such that the norm of the tail {x | C ≤ ‖f i x‖₊}.indicator (f i) is at most ε for all i. This matches Durrett's definition: lim_{M→∞} sup_i E(|X_i|; |X_i| > M) = 0.

Instances For

    On a finite measure space, the textbook definition TextbookUniformlyIntegrable' is equivalent to Mathlib's UniformIntegrable f 1 μ.

    @[reducible, inline]
    abbrev TheoryOfProbability3.UnifAbsContIntegrals {α : Type u_1} {ι : Type u_2} {m : MeasurableSpace α} (f : ια) (μ : MeasureTheory.Measure α) :

    Uniform absolute continuity of the integrals of a family f : ι → α → ℝ, expressed via Mathlib's UnifIntegrable f 1 μ.

    Instances For
      @[reducible, inline]
      abbrev TheoryOfProbability3.UnifL1Bound {α : Type u_1} {ι : Type u_2} {m : MeasurableSpace α} (f : ια) (μ : MeasureTheory.Measure α) :

      A uniform -bound on the family f : ι → α → ℝ: there exists C : ℝ≥0 with ‖f i‖₁ ≤ C for all i.

      Instances For

        On a finite measure space, the textbook notion of uniform integrability is equivalent to the conjunction of: (i) each f i is AEStronglyMeasurable, (ii) uniform absolute continuity of integrals, and (iii) a uniform -bound.

        A uniformly integrable family (in the textbook sense) is uniformly bounded in .

        A uniformly integrable family (in the textbook sense) has uniformly absolutely continuous integrals.

        Each function in a uniformly integrable family is AEStronglyMeasurable.