Lemma 4.11, p. 976 —
ProvedHarrisContact.MeanSize.lemma_4_11Let be a contact process on with death rate and birth rates , . Then for every finite ,
The expected size can decrease no faster than pure death at rate would make it. Combined with Theorem 7.6 it yields (Remark 7.12).
Formalization Note The standing hypotheses of the contact process (Harris, §2(b), p. 971; §3, p. 972) are written as binders: , , and for . Rates with never enter the model, and no hypothesis mentions them. The process is the countable-state chain on finite configurations of §4 (p. 975), with transition function the minimal solution of the backward equations for the rates (4.4)–(4.5); is an extended nonnegative real, so a divergent sum is , never . The paper states the lemma for all ; it is drafted for finite .
import Mathlib import Definitions.Def_HarrisContact_Extinction_Model open MeasureTheory Filter Topology open scoped ENNReal NNReal
namespace HarrisContact.MeanSize
/-- Harris (Ann. Probab. 2, 1974), Lemma 4.11, p. 976: `m_{t+h}(ξ) ≥ e^{−μh} m_t(ξ)` for `t, h ≥ 0`
(drafted for finite `ξ`). -/
theorem lemma_4_11 {d : ℕ} (hd : 1 ≤ d) (μ : ℝ) (lam : ℕ → ℝ) (hμ : 0 ≤ μ) (h0 : lam 0 = 0)
(hlam : ∀ k, k ≤ 2 * d → 0 ≤ lam k) :
∀ t h : ℝ, 0 ≤ t → 0 ≤ h → ∀ ξ : HarrisContact.Extinction.Config d,
ENNReal.ofReal (Real.exp (-(μ * h))) * HarrisContact.Extinction.meanSize μ lam t ξ ≤ HarrisContact.Extinction.meanSize μ lam (t + h) ξ := by sorry
end HarrisContact.MeanSize
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.