Rejection conditioning and projection preserve PMF affiliation
ProvedMinBuying.pmf_rejection_conditioningmin-buyingmtp2operations-researchpricingprobability
Let be a probability mass function on real-valued coordinates with pointwise-MTP2 real masses. If the rejection event has positive probability, then conditioning on this event and retaining the last coordinates gives an affiliated law. The law is represented by a PMF, with no finite-support assumption.
Preamble
import Definitions.Def_MinBuying_Symmetry open MeasureTheory
Formal statement
namespace MinBuying
theorem pmf_rejection_conditioning (n : ℕ) (w : PMF (Valuations (n+1)))
(hmtp : MTP2 (fun v => (w v).toReal)) (a : ℝ)
(hB : 0 < w.toMeasure {v : Valuations (n+1) | v 0 < a}) :
Affiliated (Measure.map (fun v : Valuations (n+1) => fun i : Fin n => v i.succ)
(ProbabilityTheory.cond w.toMeasure {v : Valuations (n+1) | v 0 < a})) := by
sorry
end MinBuying
Source
Titing Cui, Michael L. Hamilton, Jourdain Lamperski, and Kasra Ganjizadeh Tari, Unit-Demand Pricing for Min-Buying Customers, September 27, 2026 manuscript. https://doi.org/10.2139/ssrn.5178133 . Auxiliary affiliation and conditioning steps for the proof of Theorem 2, ec2–ec3. For association, the finite four-functions theorem is reused from Mathlib/Combinatorics/SetFamily/FourFunctions.lean.