Alphabeta Math
TheoremStatement: Literature-sourcedProof: AI-adaptedPipeline-generated
How statement and proof provenance work

The first chip identifies the source of the statement or construction; the second identifies the source of its local proof or verification.

  • Literature-sourced: the exact statement appears in a cited source; only wording and notation differ.
  • AI-adapted: a semantically identical restatement of literature-sourced material, modulo indexing, notation, and boundary cases adopted by the library.
  • AI-generated: a genuinely novel statement formulated by AI, with no source for the claim itself.

These labels describe origin, not correctness: citations and verification chips remain separate evidence.

A countable mixture dominates a sigma-finite dominated family

Statement

Assume AC. Let (Pθ)θ∈Θ be a nonempty family of probability measures dominated by a σ-finite positive measure μ on (X,A). There are a finite or countable sequence (θj) from Θ and weights wj>0 summing to one such that Q=∑jwjPθj satisfies Pθ≪Q for every θ. Repetitions are permitted if Θ is finite.

Facts & Assumptions

Given: AC, a nonempty dominated family, and a sigma-finite dominating measure.

[A1]

The Axiom of Choice: AC selects RN versions for the possibly uncountable family and countably many finite-subfamily approximations to a supremum.

[F1]

Dominated statistical experiments: every Pθ is absolutely continuous with respect to the supplied sigma-finite μ.

[F2]

Finite, sigma-finite, and semifinite measures: μ has a countable cover by measurable sets of finite measure.

[F3]

Likelihood densities and A sigma-finite signed measure that is absolutely continuous with respect to a sigma-finite positive measure has a unique almost-everywhere density: each Pθ≪λ has a nonnegative density pθ with Pθ(A)=∫Apθdλ.

[F4]

Monotone convergence for the integral: increasing nonnegative functions, including indicators and partial sums of densities, pass to their integrals.

Proof

1.1A1F1F2givenconstruct

Choose a countable measurable partition (En)n≥0 of X with μ(En)<∞ by disjointifying a sigma-finite cover [F2]. Put cn=2−(n+1)/(1+μ(En)) and λ(A)=∑n≥0cnμ(A∩En). For disjoint measurable Am, countable additivity of μ and interchange of nonnegative double series give λ(⋃mAm)=∑ncn∑mμ(Am∩En)=∑mλ(Am); the interchange is valid because both sides are the supremum of the same finite subsums. Thus λ is a measure. Also λ(X)≤∑n≥02−(n+1)=1. Since every cn>0, λ(A)=0 iff every μ(A∩En)=0, iff μ(A)=0. Thus λ is finite and equivalent to μ, so every Pθ≪λ.

2.1A1F3F4step 1.1givenchoose

For each θ, choose the RN density pθ=dPθ/dλ from [F3], finite outside a λ-null set, and let Aθ={pθ>0}. Fix θ∗; since ∫pθ∗ dλ=1, λ(Aθ∗)>0. Define s=sup⁡F⊆Θ, F finiteλ(⋃θ∈FAθ), so 0<s≤λ(X)<∞. By [A1], choose finite Fn for each n≥0 with union measure greater than s−1/(n+1). For all sufficiently large n, these unions have positive measure, so the selected index set ⋃nFn is nonempty. Set A=⋃n⋃θ∈FnAθ. Finite unions from the selected sets have measure at most s, so [F4] applied to their indicators gives λ(A)≤s by continuity from below; the approximations give λ(A)≥s. Hence λ(A)=s.

3.1F4step 2.1givencontradiction

Fix θ. If λ(Aθ∖A)>0, the increasing finite unions (⋃n≤N⋃ϕ∈FnAϕ)∪Aθ converge to A∪Aθ, whose measure is greater than s. [F4] applied to their indicators shows that some finite union then exceeds s, contradicting its definition. Therefore pθ=0 λ-almost everywhere on Ac.

4.1A1F3F4step 2.1step 3.1givenconstruct∎

Enumerate the nonempty countable collection of selected indices ⋃nFn as (θj), allowing repetitions if needed, and choose positive weights summing to one. The mixture Q=∑jwjPθj has density q=∑jwjpθj by [F4], so q>0 exactly on A up to a λ-null set. If Q(C)=0, then ∫Cq dλ=0. For each m≥1, λ(C∩A∩{q≥1/m})/m≤∫Cq dλ=0; these sets cover C∩A up to the null set where q is not positive on A. Hence λ(C∩A)=0. Also every pθ vanishes on Ac by step 3.1, so Pθ(C)=∫Cpθdλ=0. Since θ was arbitrary, Pθ≪Q for all θ.

Depends on

Used by

Dependency tree · two levels

30 results within two dependency steps of this one, each drawn at its shortest distance from it. An arrow runs from a result to what uses it, so the chart reads left to right and ends at this result, which carries a heavier outline. Every node is a link to that result. Click elsewhere on the chart to enlarge it.

Sources