Proof of Theorem 2, step (ii) — indecomposable n-step transitions
ProvedSmithHitAndRun.SymMixing.n_step_indecomposablemarkov-chainsmcmcnonperiodicityp2o-batch-pfp2bp2o-gran-per-chapterp2o-plan-paperp2o-v1
In the setting of Assumptions (a) and (b), let denote the -step transition probability. For every integer , there are no disjoint nonempty measurable sets that are both closed under :
This is the paper's nonperiodicity condition used in Theorem 2. Formalization Note is the published MarkovChainCLT.iterKernel; it starts with equal to the identity kernel.
Preamble
import Mathlib.Probability.Kernel.Invariance import Definitions.Def_MarkovIterKernel import Definitions.Def_SmithHitAndRun_SymMixing_IsSymmetricMixingKernel import Definitions.Def_SmithHitAndRun_SymMixing_IsClosedSet open MeasureTheory ProbabilityTheory
Formal statement
namespace SmithHitAndRun.SymMixing
/-- Smith (1984), proof of Theorem 2, step (ii), p. 1301: every n-step chain,
for n at least two, is indecomposable. -/
theorem n_step_indecomposable {α : Type*} [MeasurableSpace α]
(V : Measure α) [IsFiniteMeasure V] (hV : V ≠ 0)
(P : Kernel α α) [IsMarkovKernel P] (f : α → α → ENNReal)
(h : IsSymmetricMixingKernel V P f) (hpos : ∀ x y, 0 < f y x) :
∀ n : ℕ, 2 ≤ n →
¬ ∃ A₁ A₂ : Set α,
IsClosedSet (MarkovChainCLT.iterKernel P n) A₁ ∧
IsClosedSet (MarkovChainCLT.iterKernel P n) A₂ ∧ A₁ ∩ A₂ = ∅ := by sorry
end SmithHitAndRun.SymMixing
Source
Smith, Efficient Monte Carlo Procedures for Generating Points Uniformly Distributed over Bounded Regions, Oper. Res. 32(6) (1984), p. 1301, Proof of Theorem 2, condition (ii)
Human review
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.