Documentation

Mathlib.MeasureTheory.Constructions.ProjectiveFamilyContent

Additive content built from a projective family of measures #

Let P be a projective family of measures on a family of measurable spaces indexed by ι. That is, for each finite set I of indices, P I is a measure on Π j : I, α j, and for J ⊆ I, the projection from Π i : I, α i to Π i : J, α i maps P I to P J.

We build an additive content projectiveFamilyContent on the measurable cylinders, by setting projectiveFamilyContent s = P I S for s = cylinder I S, where I is a finite set of indices and S is a measurable set in Π i : I, α i.

This content will be used to define the projective limit of the family of measures P. For a countable index set and a projective family given by a sequence of kernels, the projective limit is given by the Ionescu-Tulcea theorem. For an arbitrary index set but under topological conditions on the spaces, this is the result of the Kolmogorov extension theorem. (both results are not yet in Mathlib)

Main definitions #

theorem MeasureTheory.isSetAlgebra_measurableCylinders {ι : Type u_1} {α : ιType u_2} {mα : (i : ι) → MeasurableSpace (α i)} :
theorem MeasureTheory.isSetRing_measurableCylinders {ι : Type u_1} {α : ιType u_2} {mα : (i : ι) → MeasurableSpace (α i)} :
theorem MeasureTheory.isSetSemiring_measurableCylinders {ι : Type u_1} {α : ιType u_2} {mα : (i : ι) → MeasurableSpace (α i)} :
noncomputable def MeasureTheory.projectiveFamilyFun {ι : Type u_1} {α : ιType u_2} {mα : (i : ι) → MeasurableSpace (α i)} (P : (J : Finset ι) → Measure ((j : { x : ι // x J }) → α j)) (s : Set ((i : ι) → α i)) :

For P a family of measures, with P J a measure on Π j : J, α j, we define a function projectiveFamilyFun P s by setting it to P I S if s = cylinder I S for a measurable S and to 0 if s is not a measurable cylinder.

Equations
  • One or more equations did not get rendered due to their size.
Instances For
    theorem MeasureTheory.projectiveFamilyFun_congr {ι : Type u_1} {α : ιType u_2} {mα : (i : ι) → MeasurableSpace (α i)} {P : (J : Finset ι) → Measure ((j : { x : ι // x J }) → α j)} {s : Set ((i : ι) → α i)} {I : Finset ι} {S : Set ((i : { x : ι // x I }) → α i)} (hP : IsProjectiveMeasureFamily P) (hs : s measurableCylinders α) (hs_eq : s = cylinder I S) (hS : MeasurableSet S) :
    theorem MeasureTheory.projectiveFamilyFun_empty {ι : Type u_1} {α : ιType u_2} {mα : (i : ι) → MeasurableSpace (α i)} {P : (J : Finset ι) → Measure ((j : { x : ι // x J }) → α j)} (hP : IsProjectiveMeasureFamily P) :
    theorem MeasureTheory.projectiveFamilyFun_union {ι : Type u_1} {α : ιType u_2} {mα : (i : ι) → MeasurableSpace (α i)} {P : (J : Finset ι) → Measure ((j : { x : ι // x J }) → α j)} {s t : Set ((i : ι) → α i)} (hP : IsProjectiveMeasureFamily P) (hs : s measurableCylinders α) (ht : t measurableCylinders α) (hst : Disjoint s t) :
    noncomputable def MeasureTheory.projectiveFamilyContent {ι : Type u_1} {α : ιType u_2} {mα : (i : ι) → MeasurableSpace (α i)} {P : (J : Finset ι) → Measure ((j : { x : ι // x J }) → α j)} (hP : IsProjectiveMeasureFamily P) :

    For P a projective family of measures, we define an additive content on the measurable cylinders, by setting projectiveFamilyContent s = P I S for s = cylinder I S, where I is a finite set of indices and S is a measurable set in Π i : I, α i.

    Equations
    Instances For
      theorem MeasureTheory.projectiveFamilyContent_eq {ι : Type u_1} {α : ιType u_2} {mα : (i : ι) → MeasurableSpace (α i)} {P : (J : Finset ι) → Measure ((j : { x : ι // x J }) → α j)} {s : Set ((i : ι) → α i)} (hP : IsProjectiveMeasureFamily P) :
      theorem MeasureTheory.projectiveFamilyContent_congr {ι : Type u_1} {α : ιType u_2} {mα : (i : ι) → MeasurableSpace (α i)} {P : (J : Finset ι) → Measure ((j : { x : ι // x J }) → α j)} {I : Finset ι} {S : Set ((i : { x : ι // x I }) → α i)} (hP : IsProjectiveMeasureFamily P) (s : Set ((i : ι) → α i)) (hs_eq : s = cylinder I S) (hS : MeasurableSet S) :
      (projectiveFamilyContent hP) s = (P I) S
      theorem MeasureTheory.projectiveFamilyContent_cylinder {ι : Type u_1} {α : ιType u_2} {mα : (i : ι) → MeasurableSpace (α i)} {P : (J : Finset ι) → Measure ((j : { x : ι // x J }) → α j)} {I : Finset ι} {S : Set ((i : { x : ι // x I }) → α i)} (hP : IsProjectiveMeasureFamily P) (hS : MeasurableSet S) :
      theorem MeasureTheory.projectiveFamilyContent_mono {ι : Type u_1} {α : ιType u_2} {mα : (i : ι) → MeasurableSpace (α i)} {P : (J : Finset ι) → Measure ((j : { x : ι // x J }) → α j)} {s t : Set ((i : ι) → α i)} (hP : IsProjectiveMeasureFamily P) (hs : s measurableCylinders α) (ht : t measurableCylinders α) (hst : s t) :
      theorem MeasureTheory.projectiveFamilyContent_iUnion_le {ι : Type u_1} {α : ιType u_2} {mα : (i : ι) → MeasurableSpace (α i)} {P : (J : Finset ι) → Measure ((j : { x : ι // x J }) → α j)} (hP : IsProjectiveMeasureFamily P) {s : Set ((i : ι) → α i)} (hs : ∀ (n : ), s n measurableCylinders α) (n : ) :
      (projectiveFamilyContent hP) (⋃ (i : ), ⋃ (_ : i n), s i) iFinset.range (n + 1), (projectiveFamilyContent hP) (s i)
      theorem MeasureTheory.projectiveFamilyContent_ne_top {ι : Type u_1} {α : ιType u_2} {mα : (i : ι) → MeasurableSpace (α i)} {P : (J : Finset ι) → Measure ((j : { x : ι // x J }) → α j)} {s : Set ((i : ι) → α i)} [∀ (J : Finset ι), IsFiniteMeasure (P J)] (hP : IsProjectiveMeasureFamily P) :
      theorem MeasureTheory.projectiveFamilyContent_diff {ι : Type u_1} {α : ιType u_2} {mα : (i : ι) → MeasurableSpace (α i)} {P : (J : Finset ι) → Measure ((j : { x : ι // x J }) → α j)} {s t : Set ((i : ι) → α i)} (hP : IsProjectiveMeasureFamily P) (hs : s measurableCylinders α) (ht : t measurableCylinders α) :
      theorem MeasureTheory.projectiveFamilyContent_diff_of_subset {ι : Type u_1} {α : ιType u_2} {mα : (i : ι) → MeasurableSpace (α i)} {P : (J : Finset ι) → Measure ((j : { x : ι // x J }) → α j)} {s t : Set ((i : ι) → α i)} [∀ (J : Finset ι), IsFiniteMeasure (P J)] (hP : IsProjectiveMeasureFamily P) (hs : s measurableCylinders α) (ht : t measurableCylinders α) (hts : t s) :