Theorem 8 — mean and variance of a cumulative process
ProvedSmithRegenerative.Moments.theorem_8p2o-batch-pfp2ap2o-gran-per-chapterp2o-plan-paperp2o-v1regenerative-processrenewal-rewardvariance-asymptotics
Let be a cumulative process. Write , , , , and . If the cycle length and first cycle variation are integrable, then is integrable for every and
If, in addition, the cycle length and first cycle variation have finite second moments, then has a finite second moment for every and
These formulas extend renewal-reward asymptotics from completed-cycle sums to the process observed at an arbitrary time.
Formalization Note The paper writes for ; covariance also gives the intended value when a cycle variance is zero. The limit is along real time.
Preamble
import Mathlib import Definitions.Def_SmithRegenerative_Moments_CumulativeProcess
Formal statement
namespace SmithRegenerative.Moments
open MeasureTheory Filter
/-- Smith, *Regenerative stochastic processes*, Proc. R. Soc. Lond. A
232(1188):6–31 (1955), p. 28, Theorem 8(i)–(ii), (5·3·6)–(5·3·7).
Formalization Note: `ρσ₁σ₂` is represented by covariance, preserving
within-cycle length–reward dependence and covering zero-variance cases.
Square-integrability of `w_t` is a conclusion under the second-moment
hypotheses, preventing Lean's real-variance zero default on infinite values. -/
theorem theorem_8 {Ω : Type*} [MeasurableSpace Ω]
(C : CumulativeProcess Ω)
(hμ : Integrable (C.renewal.cycleLength 1) C.renewal.P)
(hvar : Integrable (cycleVariation C.renewal C.w 1) C.renewal.P) :
((∀ t : ℝ, 0 ≤ t → Integrable (C.w t) C.renewal.P) ∧
(fun t : ℝ =>
(∫ ω, C.w t ω ∂C.renewal.P) -
C.meanReward / C.meanLength * t) =o[atTop] (fun t : ℝ => t)) ∧
(Integrable (fun ω => C.renewal.cycleLength 1 ω ^ 2) C.renewal.P →
Integrable (fun ω => cycleVariation C.renewal C.w 1 ω ^ 2) C.renewal.P →
(∀ t : ℝ, 0 ≤ t → Integrable (fun ω => C.w t ω ^ 2) C.renewal.P) ∧
(fun t : ℝ =>
ProbabilityTheory.variance (C.w t) C.renewal.P -
t / C.meanLength *
(ProbabilityTheory.variance (cycleReward C.renewal C.w 1) C.renewal.P -
2 * C.lengthRewardCovariance * (C.meanReward / C.meanLength) +
ProbabilityTheory.variance (C.renewal.cycleLength 1) C.renewal.P *
(C.meanReward / C.meanLength) ^ 2))
=o[atTop] (fun t : ℝ => t)) := by sorry
end SmithRegenerative.Moments
Source
Smith, Regenerative stochastic processes, Proc. R. Soc. Lond. A 232(1188):6–31 (1955), DOI 10.1098/rspa.1955.0198, p. 28, Theorem 8(i)–(ii), (5·3·6)–(5·3·7)
Human review
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.