Display (146), p. 889 — P*_{α,θ}(Y_n ∈ dy) = P*_{α,θ+(n−1)α}(Y_1 ∈ dy)
ProvedPoissonDirichlet.Chain.eq_146beta-distributionp2o-batch-pfp2ap2o-gran-per-chapterp2o-plan-paperp2o-v1poisson-dirichlet
Let , and . Let make independent with , and let (125). Then
The shift of parameter is what makes the transition density (139) expressible through the single family of (140).
Formalization Note. 0-based, : the law of YofR r k under equals the law of YofR r 0 under , as equality of the measures of preimages of every Borel set.
Preamble
import Mathlib import Definitions.Def_PoissonDirichlet_Chain_Setting open MeasureTheory ProbabilityTheory Filter Topology
Formal statement
namespace PoissonDirichlet.Chain
/-- (146), p. 889, with `n = k + 1`: `P*_{α,θ}(Y_n ∈ dy) = P*_{α,θ+(n-1)α}(Y_1 ∈ dy)`. Here
`μ` is `P*_{α,θ}`, `μ'` is `P*_{α,θ+kα}`, and `YofR r k` is `Y_{k+1}`. -/
theorem eq_146 (α θ : ℝ) (hα : 0 < α) (hα1 : α < 1) (hθ : -α < θ) (k : ℕ)
(μ : Measure (ℕ → ℝ)) (hμ : IsStarLaw α θ μ)
(μ' : Measure (ℕ → ℝ)) (hμ' : IsStarLaw α (θ + (k : ℝ) * α) μ') :
∀ s : Set ℝ, MeasurableSet s →
μ ((fun r => YofR r k) ⁻¹' s) = μ' ((fun r => YofR r 0) ⁻¹' s) := by sorry
end PoissonDirichlet.Chain
Source
Pitman and Yor, The two-parameter Poisson–Dirichlet distribution derived from a stable subordinator, Ann. Probab. 25 (1997), p. 889, proof of Theorem 38 (ii), (146)
Human review
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.