/- Copyright 2026 The Formal Conjectures Authors. Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with the License. You may obtain a copy of the License at https://www.apache.org/licenses/LICENSE-2.0 Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the specific language governing permissions and limitations under the License. -/ module public import FormalConjecturesUtil

Erdős Problem 1162

References:

    erdosproblems.com/1162

    [Va99] Various, Some of Paul's favorite problems. Booklet produced for the conference "Paul Erdős and his mathematics", Budapest, July 1999 (1999).

    [Py93] Pyber, L., Asymptotic results for permutation groups. Groups and computation (New Brunswick, NJ, 1991), DIMACS Ser. Discrete Math. Theoret. Comput. Sci. 11 (1993), 197-219.

    [RoTr25] Roney-Dougal, C. M. and Tracey, G., Subgroups of symmetric groups: enumeration and asymptotic properties. arXiv:2503.05416 (2025).

@[expose] public sectionopen Filter Asymptoticsnamespace Erdos1162

f n is the number of subgroups of the symmetric group $S_n$, realised as Equiv.Perm (Fin n). All subgroups are counted, not only conjugacy classes.

noncomputable def f (n : ℕ) : ℕ := Nat.card (Subgroup (Equiv.Perm (Fin n)))

f n is positive. In particular Real.log (f n) is not the junk value Real.log 0.

@[category API, AMS 5 20] theorem f_pos (n : ℕ) : 0 < f n := n:ℕ⊢ 0 < f n n:ℕ⊢ 0 < Nat.card (Subgroup (Equiv.Perm (Fin n))) All goals completed! 🐙def erdos_1162.answer : ℕ → ℝ := sorry

Give an asymptotic formula for the number of subgroups of $S_n$. Is there a statistical theorem on their order?

A problem of Erdős and Turán [Va99, 5.73]. We state the first question: erdos_1162.answer is a function asymptotically equivalent to the number $f(n)$ of subgroups of $S_n$ (an explicit formula is intended, not $f$ itself). The second question is too vague to state.

@[category research open, question value, AMS 5 20] theorem erdos_1162 : (fun n : ℕ ↦ (f n : ℝ)) ~[atTop] erdos_1162.answer := ⊢ (fun n ↦ ↑(f n)) ~[atTop] erdos_1162.answer All goals completed! 🐙

Pyber [Py93] proved that $$\log f(n)\asymp n^2.$$

@[category research solved, AMS 5 20] theorem erdos_1162.variants.pyber : (fun n : ℕ ↦ Real.log (f n : ℝ)) =Θ[atTop] fun n : ℕ ↦ (n : ℝ) ^ 2 := ⊢ (fun n ↦ Real.log ↑(f n)) =Θ[atTop] fun n ↦ ↑n ^ 2 All goals completed! 🐙

Roney-Dougal and Tracey [RoTr25] have proved that $$\log f(n)=\left(\frac{1}{16}+o(1)\right)n^2.$$ Here $\log$ is the logarithm to base $2$, as in [RoTr25].

@[category research solved, AMS 5 20] theorem erdos_1162.variants.roney_dougal_tracey : (fun n : ℕ ↦ Real.logb 2 (f n : ℝ)) ~[atTop] fun n : ℕ ↦ (n : ℝ) ^ 2 / 16 := ⊢ (fun n ↦ Real.logb 2 ↑(f n)) ~[atTop] fun n ↦ ↑n ^ 2 / 16 All goals completed! 🐙

Roney-Dougal and Tracey [RoTr25, Theorem 1] proved that there are absolute constants $\alpha>0$ and $\beta$ such that for all integers $n>1$ $$2^{n^2/16+\alpha n\log n}\leq f(n)\leq 2^{n^2/16+\beta n^{3/2}},$$ where $\log$ is the logarithm to base $2$. This implies the previous variant.

@[category research solved, AMS 5 20] theorem erdos_1162.variants.roney_dougal_tracey_bounds : ∃ α > (0 : ℝ), ∃ β : ℝ, ∀ n : ℕ, 1 < n → (2 : ℝ) ^ ((n : ℝ) ^ 2 / 16 + α * n * Real.logb 2 n) ≤ f n ∧ (f n : ℝ) ≤ (2 : ℝ) ^ ((n : ℝ) ^ 2 / 16 + β * (n : ℝ) ^ ((3 : ℝ) / 2)) := ⊢ ∃ α > 0, ∃ β, ∀ (n : ℕ), 1 < n → 2 ^ (↑n ^ 2 / 16 + α * ↑n * Real.logb 2 ↑n) ≤ ↑(f n) ∧ ↑(f n) ≤ 2 ^ (↑n ^ 2 / 16 + β * ↑n ^ (3 / 2)) All goals completed! 🐙end Erdos1162