Appendix B.4 — expansion after (37)
ProvedGenEmpLik.Coverage.expansion_after_37asymptotic-expansionempirical-likelihoodp2o-batch-pfp2bp2o-gran-per-chapterp2o-plan-paperp2o-v1
Under the conditions of Theorem 3, write , and let and be the empirical mean and variance of the influence-function observations. Then
This is the one-sided asymptotic expansion used to calibrate the confidence region.
Formalization Note Convergence in outer probability is expanded to the probabilities of events where the absolute error exceeds each positive threshold.
Preamble
import Mathlib import Definitions.Def_GenEmpLik_Coverage_AssumptionA import Definitions.Def_GenEmpLik_Coverage_AssumptionB import Definitions.Def_GenEmpLik_Coverage_optimalValue import Definitions.Def_GenEmpLik_Coverage_empiricalOptimalValue import Definitions.Def_GenEmpLik_Coverage_divergenceBall import Definitions.Def_GenEmpLik_Coverage_confidenceSet import Definitions.Def_GenEmpLik_Coverage_influence import Definitions.Def_GenEmpLik_Coverage_remainder import Definitions.Def_GenEmpLik_Coverage_robustUpper import Definitions.Def_GenEmpLik_Coverage_uniqueOptimizer import Definitions.Def_VarianceRegularization_Expansion_empMean import Definitions.Def_VarianceRegularization_Expansion_empVar open MeasureTheory ProbabilityTheory Filter Topology
Formal statement
namespace GenEmpLik.Coverage
/-- The stochastic expansion in Appendix B.4 after (37), p. 37, for T_opt. -/
theorem expansion_after_37
{Ω Ξ : Type*} [MeasurableSpace Ω] [MeasurableSpace Ξ]
(μ : Measure Ω) [IsProbabilityMeasure μ]
(ξ : ℕ → Ω → Ξ) (d : ℕ)
(X : Set (EuclideanSpace ℝ (Fin d)))
(ℓ : EuclideanSpace ℝ (Fin d) → Ξ → ℝ)
(M : Ξ → ℝ) (f : ℝ → EReal) (ρ : ℝ)
(hρ : 0 ≤ ρ) (hA : AssumptionA f)
(hX : X.Nonempty) (hB : AssumptionB X ℓ M)
(hℓmeas : ∀ x, Measurable (ℓ x))
(hM2 : Integrable (fun z => M z ^ 2) (μ.map (ξ 0)))
(hℓ2 : ∃ x0 ∈ X, Integrable (fun z => ℓ x0 z ^ 2) (μ.map (ξ 0)))
(hξmeas : ∀ i, Measurable (ξ i))
(hξindep : iIndepFun ξ μ)
(hξident : ∀ i, IdentDistrib (ξ i) (ξ 0) μ μ)
(hunique : HasUniqueOptimizer X ℓ (μ.map (ξ 0)))
(hvar : 0 < variance (ℓ (uniqueOptimizer X ℓ (μ.map (ξ 0)) hunique))
(μ.map (ξ 0))) :
∀ δ : ℝ, 0 < δ →
Tendsto (fun n : ℕ => μ {ω |
δ < |Real.sqrt (n : ℝ) *
(robustUpper X ℓ f ρ (fun i : Fin n => ξ i ω) -
optimalValue X ℓ (μ.map (ξ 0))) -
Real.sqrt (n : ℝ) *
VarianceRegularization.Expansion.empMean
(fun i : Fin n => influence ℓ
(uniqueOptimizer X ℓ (μ.map (ξ 0)) hunique)
(μ.map (ξ 0)) (ξ i ω)) -
Real.sqrt (ρ * VarianceRegularization.Expansion.empVar
(fun i : Fin n => influence ℓ
(uniqueOptimizer X ℓ (μ.map (ξ 0)) hunique)
(μ.map (ξ 0)) (ξ i ω)))|}) atTop (𝓝 0) := by sorry
end GenEmpLik.Coverage
Source
Duchi, Glynn & Namkoong, Statistics of Robust Optimization: A Generalized Empirical Likelihood Approach, arXiv:1610.03425v3, p. 37, Appendix B.4, proof of Theorem 10, display after (37)
Human review
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.