Proof of Theorem B, p. 34 — expectation of W at a threshold allocation
ProvedMultiItemRev.TwoIID.expectation_W_stepiid-goodsp2o-batch-pfp2ap2o-gran-per-chapterp2o-plan-paperp2o-v1probabilityrevenue
For independent identically distributed goods and a threshold , take the diagonal allocation and . With and ,
This evaluates the nonnegative term in equation (12) for the extreme threshold allocations.
Preamble
import Mathlib import Definitions.Def_MultiItemRev_TwoIID_Model open MeasureTheory open scoped NNReal ENNReal
Formal statement
namespace MultiItemRev.TwoIID
theorem expectation_W_step (ν : Measure ℝ≥0) [IsProbabilityMeasure ν]
(p : ℝ≥0) :
let μ := Measure.pi (fun _ : Fin 2 => ν)
(∫⁻ x, ENNReal.ofReal
(((min (x 0) (x 1) : ℝ≥0) : ℝ) * (if p ≤ x 0 then 1 else 0) -
max ((((min (x 0) (x 1) : ℝ≥0) : ℝ) - (p : ℝ))) 0) ∂μ) =
ν {t | p ≤ t} * ∫⁻ t, ((min t p : ℝ≥0) : ℝ≥0∞) ∂ν := by sorry
end MultiItemRev.TwoIID
Source
Hart and Nisan, Approximate Revenue Maximization with Multiple Items, arXiv:1204.1846v3, p. 34, proof of Theorem B, display for E[W]
Human review
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.