Lemma 7, p. 40 — lower bound for the third-phase Poisson event
ProvedResolvingNRM.FRLower.q3_probabilityconcentrationp2o-batch-pfp2bp2o-gran-per-chapterp2o-plan-paperp2o-v1poisson
Let be an integer and let have the Poisson distribution with mean . With denoting the standard normal cumulative distribution function, the third-phase event obeys
This is the second separated Poisson event used to establish a positive-probability loss scenario.
Formalization Note Equation (52) proves the displayed unrounded Gaussian-tail constant. Equation (53) and the Lemma 7 headline round upward to ; the formal statement retains the proved value. The phase length is integral as in the proof.
Preamble
import Mathlib open MeasureTheory ProbabilityTheory
Formal statement
namespace ResolvingNRM.FRLower
/-- The Q₃ component of Lemma 7 in its proved, unrounded form (52). -/
theorem q3_probability (T' : ℕ) (hT : 1 ≤ T') :
(∑' k : ℕ, if (T' : ℝ) + 6 * Real.sqrt (T' : ℝ) ≤ (k : ℝ) ∧
(k : ℝ) ≤ (T' : ℝ) + 7 * Real.sqrt (T' : ℝ) then
ProbabilityTheory.poissonPMFReal (T' : ℝ).toNNReal k else (0 : ℝ)) ≥
cdf (gaussianReal 0 1) 7 - cdf (gaussianReal 0 1) 6 -
0.9496 / Real.sqrt (T' : ℝ) := by sorry
end ResolvingNRM.FRLower
Source
Bumpensanti, Wang, A Re-solving Heuristic with Uniformly Bounded Loss for Network Revenue Management, arXiv:1802.06192v3, Lemma 7, p. 40, Q3 clause and (52)–(53)
Human review
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.