Proof of Theorem 2, p. 1215 — E = lim EXₙ⁽ⁿ⁾* = 1 − e^{−b} + a{e^{−b} − e^{−b−c}}
ProvedSamuelCahnProphet.IID.tendsto_expected_maxiidlimitp2o-batch-pfp2ap2o-gran-per-chapterp2o-plan-paperp2o-v1prophet-inequality
Fix reals , , . For each let be i.i.d., taking the values , and with probabilities , and respectively. Then
where .
This is the limit of the prophet's value along the extremal family used to show that the constant 2 of Theorem 2 cannot be lowered.
Formalization Note. The sequence is indexed by , . The law is a probability measure only once ; the finitely many earlier terms (where the Lean weights are clipped at ) do not affect the limit. The expectation is computed in and converted to a real number with toReal; it is at most once .
Preamble
import Mathlib import Definitions.Def_SamuelCahnProphet_IID_Setting open MeasureTheory Filter Topology
Formal statement
namespace SamuelCahnProphet.IID
theorem tendsto_expected_max (a b c : ℝ) (ha : 0 < a) (ha1 : a < 1) (hb : 0 < b)
(hc : 0 < c) :
Tendsto (fun k : ℕ => (Emax (threePoint ((k + 1 : ℕ) : ℝ) a b c) (k + 1)).toReal) atTop
(𝓝 (1 - Real.exp (-b) + a * (Real.exp (-b) - Real.exp (-b - c)))) := by sorry
end SamuelCahnProphet.IID
Source
Samuel-Cahn, Comparison of threshold stop rules and maximum for independent nonnegative random variables, Ann. Probab. 12 (1984), p. 1215, proof of Theorem 2, display "E = lim EXₙ⁽ⁿ⁾*"
Human review
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.