(1.4), first equality — strict-threshold reward split
ProvedSamuelCahnProphet.Median.eq_1_4_firstoptimal-stoppingp2o-batch-pfp2ap2o-gran-per-chapterp2o-plan-paperp2o-v1prophet-inequalitythreshold-rule
Let . For measurable observations , let be the first index before with , or otherwise, and let . Then
This separates the threshold level from the excess reward in (1.4).
Formalization Note The paper obtains from its nonnegative observations and median condition; the identity states just that necessary sign condition. Formalization Note Indices are zero based in Lean, so the paper's event or becomes or . Expectations are extended nonnegative integrals, and the final observation is always taken if no earlier one crosses the threshold.
Preamble
import Mathlib import Definitions.Def_SamuelCahnProphet_Median_Setting
Formal statement
namespace SamuelCahnProphet.Median
open MeasureTheory ProbabilityTheory
theorem eq_1_4_first {Ω : Type*} [MeasurableSpace Ω] (P : Measure Ω) [IsProbabilityMeasure P]
{n : ℕ} [NeZero n] (X : Fin n → Ω → ℝ) (hXm : ∀ i, Measurable (X i))
(m : ℝ) (hm : 0 ≤ m) :
EplusS P X m = ENNReal.ofReal m * P {ω | m < maxX X ω} +
∫⁻ ω, ENNReal.ofReal (X (sRule X m ω) ω - m) ∂P := by sorry
end SamuelCahnProphet.Median
Source
Samuel-Cahn, Comparison of threshold stop rules and maximum for independent nonnegative random variables, Ann. Probab. 12 (1984), p. 1214, (1.4), first equality
Human review
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.