Inequalities and asymptotics for hook lengths in ℓ \ell -regular partitions and ℓ \ell -distinct partitions
Eunmi Kim · 2025
Read Inequalities and asymptotics for hook lengths in ℓ \ell -regular partitions and ℓ \ell -distinct partitions by Eunmi Kim (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2501-10916.txt.
Introduction
Eunmi Kim studies hook lengths in two special families of number partitions in a 2025 arXiv paper. A partition splits a whole number into a sum of parts, drawn as a grid of boxes. Each box has a hook length, a count tied to its row and column. The two families bar some part sizes or repeats by a set rule.
Kim counts hooks of length one, two, and three across all partitions of a large number in each family. From sharp growth formulas for lengths two and three, she proves gap claims between the two families. The ratio of total hooks in one family to the other settles to a fixed value that rests on the rule and the length. She also proves gaps inside each family alone.
Catalog record
| CREATOR | Eunmi Kim |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 46 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2501-10916
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Inequalities and asymptotics for hook lengths in ℓ \ell -regular partitions and ℓ \ell -distinct partitions Eunmi Kim Abstract . In this article, we study hook lengths in ℓ \ell -regular partitions and ℓ \ell -distinct partitions. More precisely, we establish hook length inequalities between ℓ \ell -regular partitions and ℓ \ell -distinct partitions for hook lengths 2 2 and 3 3 , by deriving asymptotic formulas for the total number of hooks of length t t in both partition classes, for t = 1 , 2 , 3 t=1,2,3 . From these asymptotics, we show that the ratio of the total number of hooks of length t t in ℓ \ell -regular partitions to those in ℓ \ell -distinct partitions tends to a constant that depends on ℓ \ell and t t . We also provide hook length inequalities within ℓ \ell -regular partitions and within ℓ \ell -distinct partitions. 2020 Mathematics Subject Classification Primary 05A17, 11P82 1. introduction A partition \lambda=(\lambda_{1},\lambda_{2},\dots,\lambda_{k}) of a positive integer n is a non-increasing sequence of positive integers \lambda_{1}\geq\lambda_{2}\geq\dots\geq\lambda_{k} such that the parts \lambda_{j} sum up to n . A partition can be represented as a Young diagram , which is a left-justified array of square boxes. For each box v in the Young diagram of a partition \lambda , the arm length (resp. the coarm length , the leg length ) of v , denoted by a_{\lambda}(v) (resp. ca_{\lambda}(v) , l_{\lambda}(v) ), is defined as the number of boxes to the right of (resp. to the left of, below) v in the diagram of \lambda . See Figure 1 . The hook length of v is a_{\lambda}(v)+l_{\lambda}(v)+1 . In Figure 2 , the Young diagram of the partition \lambda=(5,4,2,1) with hook lengths is illustrated. m v j i Figure 1 . Arm, coarm, and leg lengths of v : a_{\lambda}(v)=j , ca_{\lambda}(v)=m , and l_{\lambda}(v)=i 8 1 6 1 3 1 1 Figure 2 . The Young diagram of the partition {(5,4,2,1)} with hook lengths Hook lengths in integer partitions play an important role in the representation theory of symmetric groups. Also, the Nekrasov-Okounkov formula [ 14 ] connects hook lengths in partitions with modular forms and q -series. Han [ 12 ] generalizes it using a combinatorial bijection [ 10 ] involving partitions, t -cores, and t -quotients. These formulas lead to an extensive study on hook lengths in partitions (see e.g. [ 3 , 4 , 7 , 8 , 9 , 11 ] ). A \ell -regular partition of a positive integer n is a partition of n in which no part is divisible by \ell and a \ell -distinct partition of a positive integer n is a partition of n in which parts appear fewer than \ell times. Glaisher bijectively proved that the number of \ell -regular partitions of n , denoted by b_{\ell}(n) , is equal to the number of \ell -distinct partitions of n , denoted by d_{\ell}(n) . Their generating functions are \sum_{n\geq 0}b_{\ell}(n)q^{n}=\prod_{\begin{subarray}{c}n\geq 1\\ \ell\nmid n\end{subarray}}\frac{1}{1-q^{n}}=\frac{(q^{\ell};q^{\ell})_{\infty}}{(q;q)_{\infty}}=\prod_{n\geq 1}\left(1+q^{n}+\cdots+q^{(\ell-1)n}\right)=\sum_{n\geq 0}d_{\ell}(n)q^{n}, where, for a\in\mathbb{C} , n\in\mathbb{N}_{0}\cup\{\infty\} , we let (a;q)_{n}:=\!\prod_{j=0}^{n-1}(1-aq^{j}) . Note that a 2 -regular partition is a partition into odd parts and that a 2 -distinct partition is a partition into distinct parts. For integers \ell\geq 2 and t\geq 1 , let b_{\ell,t}(n) (resp. d_{\ell,t}(n) ) be the total number of hooks of length t in all the \ell -regular partitions (resp. \ell -distinct partitions) of n . Andrews [ 2 , Theorem 2] proved for all n\geq 0 that the difference d_{2,1}(n)-b_{2,1}(n) is equal to the number of partitions of n such that there is exactly one part occurring three times while all other parts occur only once, which implies that d_{2,1}(n)\geq b_{2,1}(n) for all n\geq 0 . From Glaisher’s bijection, we have that for each \ell\geq 2 , b_{\ell}(n)=\sum_{t\geq 1}b_{\ell,t}(n)=\sum_{t\geq 1}d_{\ell,t}(n)=d_{\ell}(n) for all n\geq 0 . Thus, for each n\geq 0 , b_{2,t}(n)\geq d_{2,t}(n) should hold for some t\geq 2 . Ballantine, Burson, Craig, Folsom, and Wen [ 4 ] conjectured that, for t\geq 2 , there exists N_{t} such that b_{2,t}(n)\geq d_{2,t}(n) for all n>N_{t} . Craig, Dawsey, and Han [ 8 , Theorem 1.2] confirmed their conjecture by giving the asymptotics of b_{2,t}(n) and d_{2,t}(n) as n\to\infty . Li and Wang [ 13 , Theorem 1.6] proved that for n\geq 0 and \ell\geq 2 , the difference d_{\ell,1}(n)-b_{\ell,1}(n) equals the number of partitions of n in which exactly one part appears more than \ell times but fewer than 2\ell times, while all other parts occur fewer than \ell times. Thus, for \ell\geq 2 , we have d_{\ell,1}(n)\geq b_{\ell,1}(n) for all n\geq 0 . By Glaisher’s bijection, we also expect for each \ell\geq 3 that the inequality b_{\ell,t}(n)\geq d_{\ell,t}(n) holds for some t\geq 2 . In this article, for each \ell\geq 2 , we establish inequalities between b_{\ell,t}(n) and d_{\ell,t}(n) for sufficiently large integers n when t=2 and t=3 . Table 1 illustrates inequalities between b_{\ell,t}(n) and d_{\ell,t}(n) for n\gg 0 . Moreover, when t=1,2,3 , we prove that for \ell\geq 2 , the ratio \frac{d_{\ell,t}(n)}{b_{\ell,t}(n)} tends to a constant r_{\ell,t}>0 as n\to\infty . We also show that r_{\ell,t}\to 1 as \ell\to\infty , which implies that although there are inequalities between b_{\ell,t}(n) and d_{\ell,t}(n) , their ratio approaches 1 as \ell\to\infty . [formula] [formula] 1 2 3 4 5 6 \cdots 2~ d~ b~ b~ b~ b~ b~ \cdots 3~ d~ b~ b~ b? b? b? 4~ d~ d~ b~ b? 5~ d~ d~ d~ b? 6~ d~ d~ d~ b? \vdots~ \vdots~ \vdots~ \vdots~ Table 1 . For each \ell\geq 2 and t\geq 1 , the larger of b_{\ell,t}(n) and d_{\ell,t}(n) for n\gg 0 is indicated as either ‘ b ’ or ‘ d ’. The first row is confirmed by [ 8 , Theorem 1.2] , and the first column is confirmed by [ 13 , Theorem 1.6] . The second and third columns are verified by Theorem 1.1 . Additionally, ‘ b? ’ or ‘ d? ’ indicates the conjecture based on numerical data for n up to 100 . Theorem 1.1 . Let \ell\geq 2 be an integer. (1) For sufficiently large integers n , we have that \displaystyle b_{\ell,2}(n) \displaystyle\geq d_{\ell,2}(n)\quad\text{for }~2\leq\ell\leq 3, \displaystyle b_{\ell,2}(n)\leq d_{\ell,2}(n)\quad\text{for }~\ell\geq 4, \displaystyle b_{\ell,3}(n) \displaystyle\geq d_{\ell,3}(n)\quad\text{for }~2\leq\ell\leq 4, \displaystyle b_{\ell,3}(n)\leq d_{\ell,3}(n)\quad\text{for }~\ell\geq 5. (2) Furthermore, for each t=1,2,3 , there exists a constant r_{\ell,t}>0 such that as n\to\infty , \frac{d_{\ell,t}(n)}{b_{\ell,t}(n)}\to r_{\ell,t}, where we have that r_{\ell,t}\to 1 as \ell\to\infty . Theorem 1.1 follows from the asymptotic formulas of b_{\ell,t}(n) and d_{\ell,t}(n) for t=1,2,3 . Theorem 1.2 . Let \ell\geq 2 be a fixed integer. As n\to\infty , \displaystyle b_{\ell,1}(n) \displaystyle=\frac{1}{\pi}\left(1-\frac{1}{\ell}\right)\left(\frac{3}{8\ell(\ell-1)}\right)^{\frac{1}{4}}n^{-\frac{1}{4}}e^{\pi\sqrt{\frac{2n}{3}\left(1-\frac{1}{\ell}\right)}}\left(1+O\left(\frac{1}{\sqrt{n}}\right)\right), \displaystyle b_{\ell,2}(n) \displaystyle=\frac{1}{\pi}\left(1-\frac{1}{2\ell}\right)\left(\frac{3}{8\ell(\ell-1)}\right)^{\frac{1}{4}}n^{-\frac{1}{4}}e^{\pi\sqrt{\frac{2n}{3}\left(1-\frac{1}{\ell}\right)}} \displaystyle\qquad\times\left[1+\left(\frac{\pi(\ell-1)}{24}\sqrt{\frac{\ell-1}{6\ell}}-\frac{2\pi}{2\ell-1}\sqrt{\frac{\ell(\ell-1)}{6}}+\frac{1}{16\pi}\sqrt{\frac{6\ell}{\ell-1}}\right)\frac{1}{\sqrt{n}}+\mathcal{O}\left(\frac{1}{n}\right)\right], \displaystyle b_{\ell,3}(n) \displaystyle=\frac{1}{\pi}\left(1-\frac{1}{2\ell}\right)\left(\frac{3}{8\ell(\ell-1)}\right)^{\frac{1}{4}}n^{-\frac{1}{4}}e^{\pi\sqrt{\frac{2n}{3}\left(1-\frac{1}{\ell}\right)}} \displaystyle\qquad\times\left[1+\left(\frac{\pi(\ell-1)}{24}\sqrt{\frac{\ell-1}{6\ell}}-\frac{3\pi}{2\ell-1}\sqrt{\frac{\ell(\ell-1)}{6}}+\frac{1}{16\pi}\sqrt{\frac{6\ell}{\ell-1}}\right)\frac{1}{\sqrt{n}}+\mathcal{O}\left(\frac{1}{n}\right)\right]. Theorem 1.3 . Let \ell\geq 2 be a fixed integer. For t=1,2,3 , we have that as n\to\infty , d_{\ell,t}(n)=\frac{\beta_{t}(\ell)}{\pi}\left(\frac{3}{8\ell(\ell-1)}\right)^{\frac{1}{4}}n^{-\frac{1}{4}}e^{\pi\sqrt{\frac{2n}{3}\left(1-\frac{1}{\ell}\right)}}\left(1+O\left(\frac{1}{\sqrt{n}}\right)\right), where \displaystyle\beta_{1}(\ell) \displaystyle:=\frac{1}{\ell}\left[-\gamma-\psi\left(\frac{1}{\ell}\right)\right], \displaystyle\beta_{2}(\ell) \displaystyle:=\frac{1}{\ell}\left[1-\gamma-\left(1-\frac{2}{\ell}\right)\psi\left(\frac{1}{\ell}\right)-\frac{2}{\ell}\psi\left(\frac{2}{\ell}\right)\right], \displaystyle\beta_{3}(\ell) \displaystyle:=\frac{1}{\ell}\left[\frac{3}{2}-\gamma-\left(1-\frac{3}{2\ell}\right)\left(1-\frac{1}{\ell}\right)\psi\left(\frac{1}{\ell}\right)-\frac{1}{\ell}\left(1-\frac{6}{\ell}\right)\psi\left(\frac{2}{\ell}\right)-\frac{3}{2\ell}\left(1+\frac{3}{\ell}\right)\psi\left(\frac{3}{\ell}\right)\right]. Here, \gamma is the Euler-Mascheroni constant, and \psi(a):=\frac{\Gamma^{\prime}(a)}{\Gamma(a)} is the digamma function. Theorems 1.1 , 1.2 , and 1.3 are illustrated in Table 2 . We denote the main term of the asymptotic formula for b_{\ell,t}(n) in Theorem 1.2 (resp. d_{\ell,t}(n) in Theorem 1.3 ) by b^{[a]}_{\ell,t}(n) (resp. d^{[a]}_{\ell,t}(n) ). We note that the inequalities in