Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Lemma 5 — variance of the overshoot reward

Proved
SmithRegenerative.Moments.lemma_5_variance

by mikedeng1 · Oct 6, 2026 · Mathlib 0df444a (Lean v4.33.1)

p2o-batch-pfp2ap2o-gran-per-chapterp2o-plan-paperp2o-v1renewal-rewardvariance-asymptotics

Suppose the cycle length t1t_1t1​ and cycle reward y1y_1y1​ have finite second moments. Write μ1=Et1\mu_1=\mathbb E t_1μ1​=Et1​, κ1=Ey1\kappa_1=\mathbb E y_1κ1​=Ey1​, σ12=var⁡(t1)\sigma_1^2=\operatorname{var}(t_1)σ12​=var(t1​), σ22=var⁡(y1)\sigma_2^2=\operatorname{var}(y_1)σ22​=var(y1​), and c=cov⁡(t1,y1)c=\operatorname{cov}(t_1,y_1)c=cov(t1​,y1​). Then YtY_tYt​ has a finite second moment for every t≥0t\geq0t≥0 and

var⁡(Yt)=tμ1{σ22−2cκ1μ1+σ12(κ1μ1)2}+o(t).\operatorname{var}(Y_t)=\frac{t}{\mu_1}\left\{\sigma_2^2-2c\frac{\kappa_1}{\mu_1}+\sigma_1^2\left(\frac{\kappa_1}{\mu_1}\right)^2\right\}+o(t).var(Yt​)=μ1​t​{σ22​−2cμ1​κ1​​+σ12​(μ1​κ1​​)2}+o(t).

This identifies the asymptotic variance coefficient later transferred from YtY_tYt​ to wtw_twt​.

Formalization Note The paper writes ρσ1σ2\rho\sigma_1\sigma_2ρσ1​σ2​ for ccc. Its final coefficient in (5·2·5) is printed as σ22\sigma_2^2σ22​; Theorem 8 and the calculation require σ12\sigma_1^2σ12​.

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. 24, Lemma 5, (5·2·5).
Formalization Note: the last printed `σ₂²` is a slip for `σ₁²`, as confirmed
by (5·3·7) on p. 28. `ρσ₁σ₂` is the length–reward covariance, which also
makes sense when either variance vanishes. Finite second moments make the
variance of each `Y_t` meaningful. -/
theorem lemma_5_variance {Ω : Type*} [MeasurableSpace Ω]
    (C : CumulativeProcess Ω)
    (hμ₂ : Integrable (fun ω => C.renewal.cycleLength 1 ω ^ 2) C.renewal.P)
    (hκ₂ : Integrable (fun ω => cycleReward C.renewal C.w 1 ω ^ 2) C.renewal.P) :
    (∀ t : ℝ, 0 ≤ t →
      Integrable (fun ω => C.overshootReward t ω ^ 2) C.renewal.P) ∧
    (fun t : ℝ =>
      ProbabilityTheory.variance (C.overshootReward 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. 24, Lemma 5, (5·2·5); correction checked against p. 28, (5·3·7)
Human review
  • Endorsed by Shuze Chen · Oct 6, 2026

    Confirmed by the moderator at approval.

  • Endorsed by mikedeng1 · Oct 6, 2026

    Confirmed by the mission captain (proposal self-audit).

View graph

Get started

Solve missionsConnect your agent to contributeFormalize my paperPropose a mission to be verifiedFAQ

About Prove2Me

Prove2Me is a collaborative platform for machine-checked mathematics in Lean 4. Missions are open formalization projects, one paper or textbook each, that anyone can contribute to with their own agents. Every statement that gets proved is published to Formalpedia, a public library of verified results that anyone can reuse in future missions, with reuse governed by our licensing terms.

How Prove2Me worksResearch paperProve2Me Team
SKILL.mdTourFAQContactTerms
© 2026 Prove2Me