2
$\begingroup$

Let $B_{\ell,m}(x_1,x_2,\dotsc,x_{\ell-m+1})$ denote the Bell polynomials of the second kind (or say, partial Bell polynomials, (exponential) partial Bell partition polynomials). I knew that the identity \begin{equation*}\label{0Y0}\tag{o-o} \frac{B_{\ell+2m+2,\ell}\bigl(1, 0, 1, 0, 9, 0,\dotsc,[(2m-1)!!]^2, 0, [(2m+1)!!]^2\bigr)} {B_{\ell+2m,\ell}\bigl(1, 0, 1, 0, 9, 0,\dotsc,[(2m-3)!!]^2, 0, [(2m-1)!!]^2\bigr)} =(\ell+2m)^2 \end{equation*} is valid for $\ell=1,2$ and $m\in\mathbb{N}$, but not necessarily valid for $\ell\ge3$ (as the counterexample by Peter Taylor at https://mathoverflow.net/a/456761 shows).

What is the general expression of the ratio \begin{equation*} R(\ell,m)=\frac{B_{\ell+2m+2,\ell}\bigl(1, 0, 1, 0, 9, 0, 225,0,\dotsc,[(2m-1)!!]^2, 0, [(2m+1)!!]^2\bigr)} {B_{\ell+2m,\ell}\bigl(1, 0, 1, 0, 9, 0, 225,0,\dotsc,[(2m-3)!!]^2, 0, [(2m-1)!!]^2\bigr)} \end{equation*} for $\ell\ge3$ and $m\in\mathbb{N}$?

Thank you very much for your attention.

References

  1. Ch. A. Charalambides, Enumerative Combinatorics, CRC Press Series on Discrete Mathematics and its Applications. Chapman & Hall/CRC, Boca Raton, FL, 2002.
  2. L. Comtet, Advanced Combinatorics: The Art of Finite and Infinite Expansions, Revised and Enlarged Edition, D. Reidel Publishing Co., 1974; available online at https://doi.org/10.1007/978-94-010-2196-8.
  3. Feng Qi, Taylor's series expansions for real powers of two functions containing squares of inverse cosine function, closed-form formula for specific partial Bell polynomials, and series representations for real powers of Pi, Demonstratio Mathematica 55 (2022), no. 1, 710--736; available online at https://doi.org/10.1515/dema-2022-0157.
  4. Feng Qi, Explicit formulas for partial Bell polynomials, Maclaurin's series expansions of real powers of inverse (hyperbolic) cosine and sine, and series representations of powers of Pi, Research Square (2021), available online at https://doi.org/10.21203/rs.3.rs-959177/v3.
  5. Feng Qi, Gradimir V. Milovanovic, and Dongkyu Lim, Specific values of partial Bell polynomials and series expansions for real powers of functions and for composite functions, Filomat 37 (2023), no. 28, 9469--9485; available online at https://doi.org/10.2298/FIL2328469Q.
  6. Feng Qi and Peter Taylor, Series expansions for powers of sinc function and closed-form expressions for specific partial Bell polynomials, Applicable Analysis and Discrete Mathematics 18 (2024), no. 1, in press; available online at https://doi.org/10.2298/AADM230902020Q.
$\endgroup$
9
  • $\begingroup$ To be clear about which polynomials you're talking about: would $$B_{6,2}(x_1,x_2,x_3,x_4,x_5) = 6x_5x_1 + 15x_4x_2 + 10x_3^2$$ be an example of what you have in mind? $\qquad$ $\endgroup$ Commented Oct 18, 2023 at 18:21
  • $\begingroup$ @MichaelHardy Yes, you are right. \begin{equation*} B_{m,k}(x_1,x_2,\dotsc,x_{m-k+1})=\sum_{\substack{1\le j\le m-k+1\\ \ell_j\in\{0\}\cup\mathbb{N}\\ \sum_{j=1}^{m-k+1}j\ell_j=m\\ \sum_{j=1}^{m-k+1}\ell_j=k}}\frac{m!}{\prod_{j=1}^{m-k+1}\ell_j!} \prod_{j=1}^{m-k+1}\biggl(\frac{x_j}{j!}\biggr)^{\ell_j}. \end{equation*} $\endgroup$
    – qifeng618
    Commented Oct 18, 2023 at 20:39
  • 1
    $\begingroup$ Sounds like you are counting partial matchings. Writing the sums in terms of combinatorial objects instead of sums with factorials is probably a better way to go. Try colored perfect matchings with $\ell$ colors on a set of of $\ell+2m+2\ell$ elements with $\ell$ distinguished elements forced to end up with different colors inherited from edges. $\endgroup$ Commented Oct 18, 2023 at 21:17
  • $\begingroup$ @AbdelmalekAbdesselam I am not a combinatorist, so I cannot understand what you sketched. $\endgroup$
    – qifeng618
    Commented Oct 18, 2023 at 21:30
  • $\begingroup$ @AbdelmalekAbdesselam Are you combinatorially interpreting partial Bell polynomials $B_{m,k}$? My problem is to prove the identity (o-o)! Anyway, thank you very much! But I really need a complete proof of the identity (o-o) in details. $\endgroup$
    – qifeng618
    Commented Oct 19, 2023 at 2:10

2 Answers 2

1
$\begingroup$

Counterexample: consider $\ell = 3$, $m = 1$. The LHS is $$\frac{B_{7,3}({\color{red} 1}, {\color{green} 0}, {\color{blue} 1}, 0, 9)} {B_{5,3}({\color{red} 1}, {\color{green} 0}, {\color{blue} 1})} = \frac{105 \cdot {\color{green} 0}^2 \cdot {\color{blue} 1} + 70 \cdot {\color{red} 1} \cdot {\color{blue} 1}^2 + 105 \cdot {\color{red} 1} \cdot {\color{green} 0} \cdot 0 + 21 \cdot {\color{red} 1}^2 \cdot 9} {15 \cdot {\color{red} 1} \cdot {\color{green} 0}^2 + 10 \cdot {\color{red} 1}^2 \cdot {\color{blue} 1}} = \frac{259} {10}$$

$\endgroup$
1
  • $\begingroup$ Your numerical test is right, your counterexample is correct. I will modify this question very soon. Thank you very much. $\endgroup$
    – qifeng618
    Commented Oct 19, 2023 at 23:36
0
$\begingroup$

For $\ell,m\in\mathbb{N}$, the first three values of the ratio $R(\ell,m)$ are \begin{align*} R(1,m)&=(1+2m)^2,\\ R(2,m)&=(2+2m)^2,\\ R(3,m)&=(3+2m)^2\frac{\sum_{k=0}^{m+1}\frac{1}{(2k+1)^2}}{\sum_{k=0}^{m}\frac{1}{(2k+1)^2}}. \end{align*} This implies that the general expression of the ratio $R(\ell,m)$ shouldn't be of a simple form.

$\endgroup$

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .