(1.13), p. 750 — ∫_{u≤b}(e^{hu} − 1 − hu)dF ≤ ∫_{u≤t/h}(e^{hu} − 1 − hu)dF ≤ e^tβh²/2 for h ≤ t/b
ProvedNagaevLD.FukNagaev.eq_1_13large-deviationsmoment-generating-functionp2o-batch-pfp2ap2o-gran-per-chapterp2o-plan-paperp2o-v1
In the setting of Lemma 1.4 (, , , , ), suppose . Then
Together with the identity and this gives (1.10); its second inequality, which holds for every , is also used for (1.11).
Formalization Note The integrals over are expectations over the events . is assumed, as in Lemma 1.4.
Preamble
import Mathlib import Definitions.Def_NagaevLD_FukNagaev_Setting
Formal statement
namespace NagaevLD.FukNagaev
open MeasureTheory ProbabilityTheory
theorem eq_1_13 {Ω : Type*} [MeasurableSpace Ω] (P : Measure Ω) [IsProbabilityMeasure P]
(X : Ω → ℝ) (hX : Measurable X) (b : ℝ) (hb : 0 < b) (hXb : P {ω | b < X ω} = 0)
(t : ℝ) (ht : 2 ≤ t)
(hL2 : MemLp X 2 P) (h : ℝ) (hh : 0 < h) (hht : h ≤ t / b) :
∫ ω in {ω | X ω ≤ b}, (Real.exp (h * X ω) - 1 - h * X ω) ∂P ≤
∫ ω in {ω | X ω ≤ t / h}, (Real.exp (h * X ω) - 1 - h * X ω) ∂P ∧
∫ ω in {ω | X ω ≤ t / h}, (Real.exp (h * X ω) - 1 - h * X ω) ∂P ≤
Real.exp t * (∫ ω, X ω ^ 2 ∂P) * h ^ 2 / 2 := by sorry
end NagaevLD.FukNagaev
Source
Nagaev, Large deviations of sums of independent random variables, Ann. Probab. 7 (1979), p. 750, proof of Lemma 1.4, (1.13)
Human review
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.