balanced_draft_profile_variance_identity
theoremverified
The variance obtained from the squared threshold profile, the fraction of other turns, and the variance of a centered uniform variable equals the proposed balanced-draft asymptotic variance.
Statement
theorem balanced_draft_profile_variance_identity (n : ℕ) (hn : 2 ≤ n) : (16*(n:ℝ)^2/((n:ℝ)+1)^2) * (((n:ℝ)-1)/(n:ℝ)) * (1/12:ℝ) * (∫ x : ℝ in (0:ℝ)..1, (1-x) ^ ((2:ℝ)/(n:ℝ))) = balanced_draft_asymptotic_variance n
- Source
- Variance calculation in the one-agent limit of OP-1493.
- Verified
- 11 Nov 2026
- Axioms
Classical.choiceQuot.soundpropext- Built on 2
- balanced_draft_profile_square_integralFor n at least two, the integral over [0,1] of the squared limiting threshold profile (1-x)^(2/n) is n/(n+2).
- balanced_draft_asymptotic_varianceThe theoretical asymptotic variance 4n^2(n-1)/(3(n+1)^2(n+2)) of the scaled normalized utility in balanced drafting of uniform goods.
Lean source view module on GitHub
open MeasureTheory intervalIntegral theorem balanced_draft_profile_variance_identity (n : ℕ) (hn : 2 ≤ n) : (16*(n:ℝ)^2/((n:ℝ)+1)^2) * (((n:ℝ)-1)/(n:ℝ)) * (1/12:ℝ) * (∫ x : ℝ in (0:ℝ)..1, (1-x) ^ ((2:ℝ)/(n:ℝ))) = balanced_draft_asymptotic_variance n := by rw [balanced_draft_profile_square_integral n hn] unfold balanced_draft_asymptotic_variance have hnpos : (0:ℝ) < n := by exact_mod_cast (lt_of_lt_of_le (by decide : 0 < 2) hn) have hn0 : (n:ℝ) ≠ 0 := ne_of_gt hnpos have h1 : (n:ℝ)+1 ≠ 0 := by positivity have h2 : (n:ℝ)+2 ≠ 0 := by positivity field_simp ring