Lemma 6, p. 41 (PDF 42) — for c_ad < min{p², p(1−p)}: (1−p)²/2 < c^max_click(c_ad, p) < (1−p)/2
ProvedTargetGranularity.SdVsMd.lemma_6p2o-batch-pfp2ap2o-gran-per-chapterp2o-plan-paperp2o-v1targeted-advertisingthreshold-inequality
Let and , and let be the threshold of Equation (7). Then
In this range Equation (7) is in its third case, . The lemma places the md existence threshold inside the interval of click costs on which Lemma 4 controls the sd profit, which is how Lemmas 4 and 5 combine into Proposition 3.
Formalization Note is the model's standing assumption (p. 10). All denominators of the third case are positive in this range.
Preamble
import Mathlib import Definitions.Def_TargetGranularity_SdVsMd_Model
Formal statement
namespace TargetGranularity.SdVsMd
/-- Lemma 6, p. 41: for `0 < c_ad < min{p², p(1 − p)}`,
`(1 − p)²/2 < c^max_click(c_ad, p) < (1 − p)/2`. -/
theorem lemma_6 (p cad : ℝ) (hp0 : 0 < p) (hp1 : p < 1) (hcad : 0 < cad)
(hcadp2 : cad < p ^ 2) (hcadp1 : cad < p * (1 - p)) :
(1 - p) ^ 2 / 2 < cmaxClick cad p ∧ cmaxClick cad p < (1 - p) / 2 := by sorry
end TargetGranularity.SdVsMd
Source
Despotakis & Yu, Multidimensional Targeting and Consumer Response, SSRN 3900394 (version of October 11, 2022), p. 41 (PDF 42), Lemma 6
Human review
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.