Building a Markov kernel from a conditional cumulative distribution function #
Let κ : Kernel α (β × ℝ)
and ν : Kernel α β
be two finite kernels.
A function f : α × β → StieltjesFunction
is called a conditional kernel CDF of κ
with respect
to ν
if it is measurable, tends to 0 at -∞ and to 1 at +∞ for all p : α × β
,
fun b ↦ f (a, b) x
is (ν a)
-integrable for all a : α
and x : ℝ
and for all measurable
sets s : Set β
, ∫ b in s, f (a, b) x ∂(ν a) = (κ a (s ×ˢ Iic x)).toReal
.
From such a function with property hf : IsCondKernelCDF f κ ν
, we can build a Kernel (α × β) ℝ
denoted by hf.toKernel f
such that κ = ν ⊗ₖ hf.toKernel f
.
Main definitions #
Let κ : Kernel α (β × ℝ)
and ν : Kernel α β
.
ProbabilityTheory.IsCondKernelCDF
: a functionf : α × β → StieltjesFunction
is called a conditional kernel CDF ofκ
with respect toν
if it is measurable, tends to 0 at -∞ and to 1 at +∞ for allp : α × β
, iffun b ↦ f (a, b) x
is(ν a)
-integrable for alla : α
andx : ℝ
and for all measurable setss : Set β
,∫ b in s, f (a, b) x ∂(ν a) = (κ a (s ×ˢ Iic x)).toReal
.ProbabilityTheory.IsCondKernelCDF.toKernel
: from a functionf : α × β → StieltjesFunction
with the propertyhf : IsCondKernelCDF f κ ν
, build aKernel (α × β) ℝ
such thatκ = ν ⊗ₖ hf.toKernel f
.ProbabilityTheory.IsRatCondKernelCDF
: a functionf : α × β → ℚ → ℝ
is called a rational conditional kernel CDF ofκ
with respect toν
if is measurable and satisfies the same integral conditions as in the definition ofIsCondKernelCDF
, and theℚ → ℝ
functionf (a, b)
satisfies the properties of a Stieltjes function for(ν a)
-almost allb : β
.
Main statements #
ProbabilityTheory.isCondKernelCDF_stieltjesOfMeasurableRat
: iff : α × β → ℚ → ℝ
has the propertyIsRatCondKernelCDF
, thenstieltjesOfMeasurableRat f
is a functionα × β → StieltjesFunction
with the propertyIsCondKernelCDF
.ProbabilityTheory.compProd_toKernel
: forhf : IsCondKernelCDF f κ ν
,ν ⊗ₖ hf.toKernel f = κ
.
a function f : α × β → ℚ → ℝ
is called a rational conditional kernel CDF of κ
with respect
to ν
if is measurable, if fun b ↦ f (a, b) x
is (ν a)
-integrable for all a : α
and x : ℝ
and for all measurable sets s : Set β
, ∫ b in s, f (a, b) x ∂(ν a) = (κ a (s ×ˢ Iic x)).toReal
.
Also the ℚ → ℝ
function f (a, b)
should satisfy the properties of a Sieltjes function for
(ν a)
-almost all b : β
.
- measurable : Measurable f
- isRatStieltjesPoint_ae : ∀ (a : α), ∀ᵐ (b : β) ∂ν a, ProbabilityTheory.IsRatStieltjesPoint f (a, b)
- integrable : ∀ (a : α) (q : ℚ), MeasureTheory.Integrable (fun (b : β) => f (a, b) q) (ν a)
Instances For
Alias of ProbabilityTheory.setIntegral_stieltjesOfMeasurableRat_rat
.
Alias of ProbabilityTheory.setLIntegral_stieltjesOfMeasurableRat_rat
.
Alias of ProbabilityTheory.setLIntegral_stieltjesOfMeasurableRat
.
Alias of ProbabilityTheory.setIntegral_stieltjesOfMeasurableRat
.
This property implies IsRatCondKernelCDF
. The measurability, integrability and integral
conditions are the same, but the limit properties of IsRatCondKernelCDF
are replaced by
limits of integrals.
- measurable : Measurable f
- tendsto_integral_of_antitone : ∀ (a : α) (seq : ℕ → ℚ), Antitone seq → Filter.Tendsto seq Filter.atTop Filter.atBot → Filter.Tendsto (fun (m : ℕ) => ∫ (c : β), f (a, c) (seq m) ∂ν a) Filter.atTop (nhds 0)
- tendsto_integral_of_monotone : ∀ (a : α) (seq : ℕ → ℚ), Monotone seq → Filter.Tendsto seq Filter.atTop Filter.atTop → Filter.Tendsto (fun (m : ℕ) => ∫ (c : β), f (a, c) (seq m) ∂ν a) Filter.atTop (nhds ((ν a) Set.univ).toReal)
- integrable : ∀ (a : α) (q : ℚ), MeasureTheory.Integrable (fun (c : β) => f (a, c) q) (ν a)
Instances For
Alias of ProbabilityTheory.IsRatCondKernelCDFAux.setIntegral_iInf_rat_gt
.
A function f : α × β → StieltjesFunction
is called a conditional kernel CDF of κ
with
respect to ν
if it is measurable, tends to 0 at -∞ and to 1 at +∞ for all p : α × β
,
fun b ↦ f (a, b) x
is (ν a)
-integrable for all a : α
and x : ℝ
and for all
measurable sets s : Set β
, ∫ b in s, f (a, b) x ∂(ν a) = (κ a (s ×ˢ Iic x)).toReal
.
- measurable : ∀ (x : ℝ), Measurable fun (p : α × β) => ↑(f p) x
- integrable : ∀ (a : α) (x : ℝ), MeasureTheory.Integrable (fun (b : β) => ↑(f (a, b)) x) (ν a)
- tendsto_atTop_one : ∀ (p : α × β), Filter.Tendsto (↑(f p)) Filter.atTop (nhds 1)
- tendsto_atBot_zero : ∀ (p : α × β), Filter.Tendsto (↑(f p)) Filter.atBot (nhds 0)
Instances For
A measurable function α → StieltjesFunction
with limits 0 at -∞ and 1 at +∞ gives a measurable
function α → Measure ℝ
by taking StieltjesFunction.measure
at each point.
A function f : α × β → StieltjesFunction
with the property IsCondKernelCDF f κ ν
gives a
Markov kernel from α × β
to ℝ
, by taking for each p : α × β
the measure defined by f p
.
Equations
- ProbabilityTheory.IsCondKernelCDF.toKernel f hf = { toFun := fun (p : α × β) => (f p).measure, measurable' := ⋯ }
Instances For
Equations
- ⋯ = ⋯