# The number of $n$-tuples of nonnegative integers $(x_1,\ldots,x_n)$ satisfying the equation $x_1+\cdots+x_n = r$

For given positive integers $r,v,n$ let $S(r,v,n)$ denote the number of $n$-tuples of nonnegative integers $(x_1,\ldots,x_n)$ satisfying the equation $x_1+\cdots+x_n = r$ and such that $x_i \leq v$ for $i = 1,\ldots,n$. Prove that $$S(r,v,n) = \sum_{k=0}^m (-1)^k \binom{n}{k} \binom{r-(v+1)k+n-1}{n-1},$$ where $m = \min\left\{n,\left[\frac{r}{v+1}\right]\right\}$.

If $v \geq r$, then we see that the sum is just $\binom{n+r-1}{n-1}$, which is the number of nonnegative integer solutions to $x_1+\cdots+x_n = r$. If $v < r$, then how do we get the formula? How is $v+1$ involved?

We have from first principles that this value is

$$[z^r] (1+z+z^2+z^3+\cdots+z^v)^n = [z^r] \frac{(1-z^{v+1})^n}{(1-z)^n}.$$

This is

$$[z^r] \frac{1}{(1-z)^n} \sum_{k=0}^n {n\choose k} (-1)^k z^{(v+1)k} \\ = \sum_{k=0}^n {n\choose k} (-1)^k [z^{r-(v+1)k}] \frac{1}{(1-z)^n} \\ = \sum_{k=0}^n {n\choose k} (-1)^k {n-1+r-(v+1)k\choose n-1}.$$

The claim now follows.

We get for the upper limit that we must have $n-1+r-(v+1)k \ge n-1$ or $r/(v+1)\ge k.$ The first binomial coefficient adds an upper limit of $n$ so that we obtain

$$\min(n, \lfloor r/(v+1) \rfloor).$$

• Nice answer (+1) Jul 23, 2016 at 22:13
• Thanks ever so much. This answer is even nicer. Unfortunately not many people will see it since it has been put on hold. Jul 23, 2016 at 22:18
• @MarkusScheuer Never mind my previous comment, it really is a duplicate. Jul 23, 2016 at 22:42

$\newcommand{\angles}[1]{\left\langle\,{#1}\,\right\rangle} \newcommand{\braces}[1]{\left\lbrace\,{#1}\,\right\rbrace} \newcommand{\bracks}[1]{\left\lbrack\,{#1}\,\right\rbrack} \newcommand{\dd}{\mathrm{d}} \newcommand{\ds}[1]{\displaystyle{#1}} \newcommand{\expo}[1]{\,\mathrm{e}^{#1}\,} \newcommand{\half}{{1 \over 2}} \newcommand{\ic}{\mathrm{i}} \newcommand{\iff}{\Longleftrightarrow} \newcommand{\imp}{\Longrightarrow} \newcommand{\Li}[1]{\,\mathrm{Li}_{#1}} \newcommand{\ol}[1]{\overline{#1}} \newcommand{\pars}[1]{\left(\,{#1}\,\right)} \newcommand{\partiald}[3][]{\frac{\partial^{#1} #2}{\partial #3^{#1}}} \newcommand{\ul}[1]{\underline{#1}} \newcommand{\root}[2][]{\,\sqrt[#1]{\,{#2}\,}\,} \newcommand{\totald}[3][]{\frac{\mathrm{d}^{#1} #2}{\mathrm{d} #3^{#1}}} \newcommand{\verts}[1]{\left\vert\,{#1}\,\right\vert}$ \begin{align} &\color{#f00}{\sum_{x_{1}\ =\ 0}^{\nu}\cdots\sum_{x_{n}\ =\ 0}^{\nu}\,\, \delta_{\ds{x_{1} + \cdots + x_{n},r}}} = \sum_{x_{1}\ =\ 0}^{\nu}\cdots\sum_{x_{n}\ =\ 0}^{\nu} \,\,\oint_{\verts{z}\ =\ 1^{-}}\,\,\, {1 \over z^{r\ +\ 1\ -\ x_{1}\ -\ \cdots\ -\ x_{n}}} \,\,\,\,\,{\dd z \over 2\pi\ic} \\[4mm] = &\ \oint_{\verts{z} = 1^{-}}\,\,{1 \over z^{r + 1}}\,\,\, \pars{\sum_{x\ =\ 0}^{\nu}z^{x}}^{n}\,{\dd z \over 2\pi\ic} = \oint_{\verts{z} = 1^{-}}\,\,{1 \over z^{r + 1}}\,\,\, \pars{z^{\nu + 1} - 1 \over z - 1}^{n}\,{\dd z \over 2\pi\ic} \\[4mm] = &\ \oint_{\verts{z} = 1^{-}}\,\,\,{1 \over z^{r + 1}}\,\,\, \sum_{\ell = 0}^{n}{n \choose \ell}\pars{-x^{\nu + 1}}^{\ell} \sum_{\ell' = n}^{\infty}{n \choose \ell'}\pars{-z}^{\ell'}\,{\dd z \over 2\pi\ic} \\[4mm] = &\ \sum_{\ell = 0}^{n}{n \choose \ell}\pars{-1}^{\ell} \sum_{\ell' = 0}^{\infty}{n \choose \ell'}\pars{-1}^{\ell'} \oint_{\verts{z} = 1^{-}}\,\,\, {1 \over z^{r - \nu\ell -\ell -\ell '+ 1}}\,\,\, \,{\dd z \over 2\pi\ic} \\[4mm] = &\ \sum_{\ell = 0}^{n}{n \choose \ell}\pars{-1}^{\ell}\,\,\sum_{\ell' = 0}^{\infty} \pars{-1}^{\ell'}{n \choose \ell'}\delta_{r - \nu\ell - \ell - \ell' + 1,1} \\[4mm] = &\ \left.\sum_{\ell = 0}^{n}{n \choose \ell}\pars{-1}^{\ell} \pars{-1}^{r - \nu\ell -\ell}\,\,{n \choose r - \nu\ell - \ell} \,\right\vert_{\ r\ -\ \nu\ell\ -\ \ell\ \geq\ 0} \\[4mm] = &\ \color{#f00}{\pars{-1}^{r}\sum_{\ell = 0}^{m}\pars{-1}^{\nu\ell}\,\, {n \choose \ell}{n \choose r - \nu\ell - \ell}}\quad \mbox{where}\quad \color{#f00}{m} \equiv \color{#f00}{\min\braces{n,\left\lfloor{r \over \nu + 1}\right\rfloor}} \end{align}

This is inclusion exclusion. The $k = 0$ term corresponds to all non-negative solutions to the equation, the $k = 1$ term corresponds to all non-negative solutions where at least 1 of the $x_i$ is larger than $v$, the $k = 2$ term corresponds to all non-negative solutions where at least 2 of the $x_i$ are larger than $v$, and so on.