Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Eqs. (1.3.2)-(1.3.3) — the mean-value formulas

Proved
PalmQueueing.Palm.mean_value_formulas

by mikedeng1 · Oct 1, 2026 · Mathlib 0df444a (Lean v4.33.1)

palm-calculuspoint-processqueueingstationary-process

Let (N,θt,P)(N, \theta_t, P)(N,θt​,P) be a stationary point process with finite intensity and PN0P^0_NPN0​ the associated Palm probability. Let {Zt}\{Z_t\}{Zt​}, t∈Rt \in \mathbb{R}t∈R, be a stochastic process with values in a measurable space (K,K)(K, \mathcal{K})(K,K) such that

Zt=Z0∘θt.(1.3.1)Z_t = Z_0 \circ \theta_t . \tag{1.3.1}Zt​=Z0​∘θt​.(1.3.1)

Then for all non-negative measurable functions g:(K,K)→(R,B)g : (K, \mathcal{K}) \to (\mathbb{R}, \mathcal{B})g:(K,K)→(R,B),

E[g(Z0)]  =  EN0[∫0T1g(Zt) dt]EN0[T1](1.3.2)E\big[g(Z_0)\big] \;=\; \frac{E^0_N\Big[\int_0^{T_1} g(Z_t)\,dt\Big]}{E^0_N[T_1]} \tag{1.3.2}E[g(Z0​)]=EN0​[T1​]EN0​[∫0T1​​g(Zt​)dt]​(1.3.2)

and

EN0[g(Z0)]  =  E[∑n∈Zg(ZTn)1{Tn∈(0,1]}]E[∑n∈Z1{Tn∈(0,1]}].(1.3.3)E^0_N\big[g(Z_0)\big] \;=\; \frac{E\Big[\sum_{n \in \mathbb{Z}} g(Z_{T_n}) \mathbf{1}_{\{T_n \in (0,1]\}}\Big]} {E\Big[\sum_{n \in \mathbb{Z}} \mathbf{1}_{\{T_n \in (0,1]\}}\Big]} . \tag{1.3.3}EN0​[g(Z0​)]=E[∑n∈Z​1{Tn​∈(0,1]}​]E[∑n∈Z​g(ZTn​​)1{Tn​∈(0,1]}​]​.(1.3.3)

The book notes that these "just rephrase the inversion formula (1.2.25) and the definition formula (1.2.1) of PN0P^0_NPN0​", and that (1.3.2) is well known in the theory of renewal and regenerative processes.

Both denominators are non-zero and finite: EN0[T1]=1/λE^0_N[T_1] = 1/\lambdaEN0​[T1​]=1/λ by (1.2.27), and the denominator of (1.3.3) is λ\lambdaλ itself.

Preamble
import Mathlib
import Definitions.Def_PalmQueueing_Palm_PointProcess

/-!
# Eqs. (1.3.2) and (1.3.3): the mean-value formulas (§1.3.1, p.21)
-/
Formal statement
namespace PalmQueueing.Palm

open MeasureTheory

variable {Ω : Type*} [MeasurableSpace Ω]

/-- **The mean-value formulas**, Eqs. (1.3.2) and (1.3.3) (§1.3.1, p.21). Let `{Z_t}` take values
in a measurable space `(K, 𝒦)` and satisfy `(1.3.1) Z_t = Z₀ ∘ θ_t`. Then for every non-negative
measurable `g : (K, 𝒦) → (ℝ, B)`, the stationary and Palm expectations of `g(Z₀)` each express the
other as a ratio:

`(1.3.2)  E[g(Z₀)] = E⁰_N[ ∫_0^{T₁} g(Z_t) dt ] / E⁰_N[T₁]`,

`(1.3.3)  E⁰_N[g(Z₀)] = E[ Σ_{n ∈ ℤ} g(Z_{T_n}) 1_{T_n ∈ (0,1]} ] / E[ Σ_{n ∈ ℤ} 1_{T_n ∈ (0,1]} ]`.

The book records that these "just rephrase the inversion formula (1.2.25) and the definition
formula (1.2.1) of `P⁰_N`", and that (1.3.2) is the familiar renewal- and regenerative-process
identity. Both denominators are non-zero and finite: `E⁰_N[T₁] = 1/λ` by (1.2.27) and the second is
`λ` itself. -/
theorem mean_value_formulas {K : Type*} [MeasurableSpace K] (S : PalmSetting Ω)
    (Z : ℝ → Ω → K) (hZmeas : ∀ t, Measurable (Z t))
    (hZ : ∀ (t : ℝ) (ω : Ω), Z t ω = Z 0 (S.θ t ω))
    (g : K → ENNReal) (hg : Measurable g) :
    (∫⁻ ω, g (Z 0 ω) ∂S.P
        = (∫⁻ ω, ∫⁻ t in Set.Ioc (0 : ℝ) (S.N.T 1 ω), g (Z t ω) ∂(volume : Measure ℝ) ∂S.P0)
            / (∫⁻ ω, ENNReal.ofReal (S.N.T 1 ω) ∂S.P0))
      ∧ (∫⁻ ω, g (Z 0 ω) ∂S.P0
        = (∫⁻ ω, ∑' n : ℤ,
              Set.indicator (Set.Ioc (0 : ℝ) 1) (fun _ => g (Z (S.N.T n ω) ω)) (S.N.T n ω) ∂S.P)
            / (∫⁻ ω, ∑' n : ℤ,
              Set.indicator (Set.Ioc (0 : ℝ) 1) (fun _ => (1 : ENNReal)) (S.N.T n ω) ∂S.P)) := by sorry

end PalmQueueing.Palm
Source
Baccelli & Bremaud, Elements of Queueing Theory: Palm Martingale Calculus and Stochastic Recurrences, 2nd ed., Springer 2003, p. 21, §1.3.1, Eqs. (1.3.2)-(1.3.3)
Human review
  • Endorsed by Shuze Chen · Oct 5, 2026

    Confirmed by the moderator at approval.

  • Endorsed by mikedeng1 · Oct 5, 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