Proposition C.4.9 — a conforming AS has π_i(N) → π_i for all i and constant average costs J(N) → J_R
ProvedSennottDP.ChainASM.conforming_steady_state_avg_costapproximating-sequenceaverage-costmarkov-chainp2o-batch-b23ap2o-gran-per-chapterp2o-plan-bookp2o-v1
Let be a standard Markov chain with costs on a denumerable state space , and let be a conforming approximating sequence. Let be the positive recurrent class of containing and its average cost. Then:
- for all ;
- , where is the constant average cost on for :
Conformity thus delivers exactly what an approximation scheme is used for: the steady state distribution and the average cost of the finite chains converge to those of .
Formalization Note The positive recurrent class containing is written as the communicating class of (for a standard chain it is positive recurrent, Proposition C.2.6). Part 2 is stated as the existence of a sequence that equals for every for all sufficiently large and converges to .
Preamble
import Mathlib import Definitions.Def_SennottDP_ChainASM_MarkovChain import Definitions.Def_SennottDP_ChainASM_ApproxSeq open scoped ENNReal NNReal open Filter Topology open Classical
Formal statement
namespace SennottDP.ChainASM
theorem conforming_steady_state_avg_cost {S : Type*} [Countable S] [Infinite S]
(Γ : MC S) (z : S) (hz : Γ.IsZStandard z) (AS : ApproxSeq Γ)
(hconf : AS.IsConforming z) :
(∀ i : S, Tendsto (fun N => AS.steadyStateN N i) atTop (𝓝 (Γ.steadyState i))) ∧
∃ Jc : ℕ → ℝ≥0∞, (∀ᶠ N in atTop, ∀ i ∈ AS.SN N, AS.avgCostN N i = Jc N) ∧
Tendsto Jc atTop (𝓝 (Γ.classAvgCost (Γ.commClass z))) := by sorry
end SennottDP.ChainASM
Source
Sennott, Stochastic Dynamic Programming and the Control of Queueing Systems (Wiley, 1999), p. 307, Proposition C.4.9
Human review
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.