Documentation

Mathlib.Data.Nat.Multiplicity

Natural number multiplicity #

This file contains lemmas about the multiplicity function (the maximum prime power dividing a number) when applied to naturals, in particular calculating it for factorials and binomial coefficients.

Multiplicity calculations #

Other declarations #

Tags #

Legendre, p-adic

theorem Nat.multiplicity_eq_card_pow_dvd {m : ℕ} {n : ℕ} {b : ℕ} (hm : m ≠ 1) (hn : 0 < n) (hb : Nat.log m n < b) :
multiplicity m n = ↑(Finset.filter (fun (i : ℕ) => m ^ i ∣ n) (Finset.Ico 1 b)).card

The multiplicity of m in n is the number of positive natural numbers i such that m ^ i divides n. This set is expressed by filtering Ico 1 b where b is any bound greater than log m n.

theorem Nat.Prime.multiplicity_mul {p : ℕ} {m : ℕ} {n : ℕ} (hp : Nat.Prime p) :
theorem Nat.Prime.multiplicity_pow {p : ℕ} {m : ℕ} {n : ℕ} (hp : Nat.Prime p) :
theorem Nat.Prime.multiplicity_pow_self {p : ℕ} {n : ℕ} (hp : Nat.Prime p) :
multiplicity p (p ^ n) = ↑n
theorem Nat.Prime.multiplicity_factorial {p : ℕ} (hp : Nat.Prime p) {n : ℕ} {b : ℕ} :
Nat.log p n < b → multiplicity p n.factorial = ↑(∑ i ∈ Finset.Ico 1 b, n / p ^ i)

Legendre's Theorem

The multiplicity of a prime in n! is the sum of the quotients n / p ^ i. This sum is expressed over the finset Ico 1 b where b is any bound greater than log p n.

theorem Nat.Prime.sub_one_mul_multiplicity_factorial {n : ℕ} {p : ℕ} (hp : Nat.Prime p) :
(p - 1) * (multiplicity p n.factorial).get ⋯ = n - (p.digits n).sum

For a prime number p, taking (p - 1) times the multiplicity of p in n! equals n minus the sum of base p digits of n.

theorem Nat.Prime.multiplicity_factorial_mul_succ {n : ℕ} {p : ℕ} (hp : Nat.Prime p) :
multiplicity p (p * (n + 1)).factorial = multiplicity p (p * n).factorial + multiplicity p (n + 1) + 1

The multiplicity of p in (p * (n + 1))! is one more than the sum of the multiplicities of p in (p * n)! and n + 1.

theorem Nat.Prime.multiplicity_factorial_mul {n : ℕ} {p : ℕ} (hp : Nat.Prime p) :
multiplicity p (p * n).factorial = multiplicity p n.factorial + ↑n

The multiplicity of p in (p * n)! is n more than that of n!.

theorem Nat.Prime.pow_dvd_factorial_iff {p : ℕ} {n : ℕ} {r : ℕ} {b : ℕ} (hp : Nat.Prime p) (hbn : Nat.log p n < b) :
p ^ r ∣ n.factorial ↔ r ≤ ∑ i ∈ Finset.Ico 1 b, n / p ^ i

A prime power divides n! iff it is at most the sum of the quotients n / p ^ i. This sum is expressed over the set Ico 1 b where b is any bound greater than log p n

theorem Nat.Prime.multiplicity_factorial_le_div_pred {p : ℕ} (hp : Nat.Prime p) (n : ℕ) :
multiplicity p n.factorial ≤ ↑(n / (p - 1))
theorem Nat.Prime.multiplicity_choose_aux {p : ℕ} {n : ℕ} {b : ℕ} {k : ℕ} (hp : Nat.Prime p) (hkn : k ≤ n) :
∑ i ∈ Finset.Ico 1 b, n / p ^ i = ∑ i ∈ Finset.Ico 1 b, k / p ^ i + ∑ i ∈ Finset.Ico 1 b, (n - k) / p ^ i + (Finset.filter (fun (i : ℕ) => p ^ i ≤ k % p ^ i + (n - k) % p ^ i) (Finset.Ico 1 b)).card
theorem Nat.Prime.multiplicity_choose' {p : ℕ} {n : ℕ} {k : ℕ} {b : ℕ} (hp : Nat.Prime p) (hnb : Nat.log p (n + k) < b) :
multiplicity p ((n + k).choose k) = ↑(Finset.filter (fun (i : ℕ) => p ^ i ≤ k % p ^ i + n % p ^ i) (Finset.Ico 1 b)).card

The multiplicity of p in choose (n + k) k is the number of carries when k and n are added in base p. The set is expressed by filtering Ico 1 b where b is any bound greater than log p (n + k).

theorem Nat.Prime.multiplicity_choose {p : ℕ} {n : ℕ} {k : ℕ} {b : ℕ} (hp : Nat.Prime p) (hkn : k ≤ n) (hnb : Nat.log p n < b) :
multiplicity p (n.choose k) = ↑(Finset.filter (fun (i : ℕ) => p ^ i ≤ k % p ^ i + (n - k) % p ^ i) (Finset.Ico 1 b)).card

The multiplicity of p in choose n k is the number of carries when k and n - k are added in base p. The set is expressed by filtering Ico 1 b where b is any bound greater than log p n.

A lower bound on the multiplicity of p in choose n k.

theorem Nat.Prime.multiplicity_choose_prime_pow_add_multiplicity {p : ℕ} {n : ℕ} {k : ℕ} (hp : Nat.Prime p) (hkn : k ≤ p ^ n) (hk0 : k ≠ 0) :
multiplicity p ((p ^ n).choose k) + multiplicity p k = ↑n
theorem Nat.Prime.multiplicity_choose_prime_pow {p : ℕ} {n : ℕ} {k : ℕ} (hp : Nat.Prime p) (hkn : k ≤ p ^ n) (hk0 : k ≠ 0) :
multiplicity p ((p ^ n).choose k) = ↑(n - (multiplicity p k).get ⋯)
theorem Nat.Prime.dvd_choose_pow {p : ℕ} {n : ℕ} {k : ℕ} (hp : Nat.Prime p) (hk : k ≠ 0) (hkp : k ≠ p ^ n) :
p ∣ (p ^ n).choose k
theorem Nat.Prime.dvd_choose_pow_iff {p : ℕ} {n : ℕ} {k : ℕ} (hp : Nat.Prime p) :
p ∣ (p ^ n).choose k ↔ k ≠ 0 ∧ k ≠ p ^ n
theorem Nat.multiplicity_two_factorial_lt {n : ℕ} :
n ≠ 0 → multiplicity 2 n.factorial < ↑n