§1B, proof of Proposition 1 — variance as a pairwise sum
ProvedDiaconisStroock.Poincare.var_eq_half_sum_sqp2o-batch-pfp2ap2o-gran-per-chapterp2o-plan-paperp2o-v1poincare-inequalityprobabilityvariance
Let be a probability distribution on a finite set and . Its variance equals half the average squared difference between two independent draws from :
This is the opening identity in the proof of Proposition 1 and connects the variance to ordered pairs of states.
Formalization Note No transition matrix is needed for this identity; only the probability distribution is assumed.
Preamble
import Mathlib import Definitions.Def_mm_lower
Formal statement
namespace DiaconisStroock.Poincare
open MarkovMixing
open scoped BigOperators
/-- The first equality in the proof of Proposition 1, p. 38. -/
theorem var_eq_half_sum_sq {V : Type*} [Fintype V] [DecidableEq V]
(π : V → ℝ) (hπ : IsDist π) (φ : V → ℝ) :
distVar π φ = 2⁻¹ * ∑ x, ∑ y, (φ x - φ y) ^ 2 * (π x * π y) := by sorry
end DiaconisStroock.Poincare
Source
Diaconis and Stroock, Geometric bounds for eigenvalues of Markov chains, Ann. Appl. Probab. 1 (1991), p. 38, §1B, proof of Proposition 1, first display, https://doi.org/10.1214/aoap/1177005980
Human review
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.