§2, p. 11 — absent information the firm exploits iff p ≥ p̂ = (v_S − v_N)/(v_W − v_N)
ProvedExpBias.OwnBias.pHat_thresholddecision-theoryinnovationp2o-batch-pfp2ap2o-gran-per-chapterp2o-plan-paperp2o-v1threshold
Under the standing assumptions (, , ), the no-information threshold lies strictly between and , and exploiting without any experiment is at least as good as the status quo exactly when the prior is at least :
The threshold separates the firms predisposed to exploit from those predisposed not to; Proposition 1 shows it also decides which biased experiment the firm prefers.
Preamble
import Mathlib import Definitions.Def_ExpBias_OwnBias_Model
Formal statement
namespace ExpBias.OwnBias
theorem pHat_threshold (P : Params) (hP : P.Standing) :
0 < pHat P ∧ pHat P < 1 ∧
(P.vS ≤ P.p * P.vW + (1 - P.p) * P.vN ↔ pHat P ≤ P.p) := by sorry
end ExpBias.OwnBias
Source
Gans, A Synthetic Model of Disruption and Experimentation, NBER Working Paper 29091 (July 2021), p. 11 (PDF 12), §2
Human review
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.