sum antidiagonal div factorial
ProvedLevyStochCalc.Probability.sum_antidiagonal_div_factorialcontext-reviewlevystochcalcsource-import
The source theorem LevyStochCalc.Probability.sum_antidiagonal_div_factorial is extracted from LevyStochCalc.Probability.HermiteGenerating. The original Lean proposition is retained verbatim, and its proof is tracked separately for remote certification.
Preamble
import Definitions.Def_LevyStochCalc_Probability_HermiteGenerating_F001 /- Copyright (c) 2026 Christian Garry. All rights reserved. Released under Apache 2.0 license as described in the file LICENSE. Authors: Christian Garry -/ /-! # The generating function of the Hermite polynomials of variance `τ` The Hermite polynomials of variance `τ` are generated over `ℂ` by `e^(t x - t ^ 2 τ / 2) = ∑_r (t ^ r / r!) H_r(x; τ)`, which follows from the Appell binomial decomposition `H_n(x; τ) / n! = ∑_(a + m = n) (x ^ a / a!) (H_m(0; τ) / m!)` by an antidiagonal Cauchy product of the exponential series with the series of the values at the origin, the latter summing to `e^(-t ^ 2 τ / 2)` because `H_(2b)(0; τ) = (-τ / 2) ^ b (2b)! / b!` and `H_(2b+1)(0; τ) = 0`. A complex scalar covers the purely imaginary frequencies `t = i a`, where the identity reads `e^(i a x + a ^ 2 τ / 2) = ∑_r ((i a) ^ r / r!) H_r(x; τ)`. Both factors of the Cauchy product are absolutely summable, hence so is the scaled Hermite series itself, at every argument and every complex scalar. ## Main statements * `LevyStochCalc.Probability.hermiteScaled_two_mul_zero` — `H_(2b)(0; τ) = (-τ / 2) ^ b (2b)! / b!`. * `LevyStochCalc.Probability.hermiteScaled_two_mul_add_one_zero` — `H_(2b+1)(0; τ) = 0`. * `LevyStochCalc.Probability.sum_antidiagonal_hermiteScaled` — the Appell binomial identity `H_n(x; τ) / n! = ∑_(a + m = n) (x ^ a / a!) (H_m(0; τ) / m!)`. * `LevyStochCalc.Probability.hasSum_hermiteScaled_zero` — `e^(-t ^ 2 τ / 2) = ∑_m (t ^ m / m!) H_m(0; τ)`. * `LevyStochCalc.Probability.hasSum_hermiteScaled` — `e^(t x - t ^ 2 τ / 2) = ∑_r (t ^ r / r!) H_r(x; τ)` for `t : ℂ`. * `LevyStochCalc.Probability.summable_norm_expTerm` — absolute summability of `z ^ a / a!`. * `LevyStochCalc.Probability.summable_norm_hermiteScaled_zero` — absolute summability of `t ^ m / m! · H_m(0; τ)`. * `LevyStochCalc.Probability.summable_norm_hermiteScaled` — absolute summability of `t ^ r / r! · H_r(x; τ)` at every argument. * `LevyStochCalc.Probability.hasSum_hermiteScaled_real` — the same identity for a real scalar. * `LevyStochCalc.Probability.hasSum_hermiteScaled_I` — `e^(i a x + a ^ 2 τ / 2) = ∑_r ((i a) ^ r / r!) H_r(x; τ)`. * `LevyStochCalc.Probability.sum_antidiagonal_div_factorial` — `∑_(r + s = d) A ^ r B ^ s / (r! s!) = (A + B) ^ d / d!`. -/ namespace LevyStochCalc.Probability open Finset
Formal statement
theorem _root_.LevyStochCalc.Probability.sum_antidiagonal_div_factorial {K : Type*} [Field K] [CharZero K] (A B : K) (d : ℕ) :
∑ p ∈ antidiagonal d, A ^ p.1 * B ^ p.2 / ((p.1.factorial : K) * (p.2.factorial : K))
= (A + B) ^ d / (d.factorial : K) := by sorry
Source