Show that $\sum_{k=0}^{n}\binom{n}{k}^2=\frac{n+1}{n}\sum_{k=1}^{n}\binom{n}{k}\binom{n}{k-1}$

Question

I came across this result while trying to solve this:

My proof is cumbersome, so I hope that someone can come up with a more elegant proof.

Note: I know that $\sum_{k=0}^{n}\binom{n}{k}^2 =\binom{2n}{n} $.

score 4 · Accepted Answer · answered Sep 29 '16 at 22:02

The fact that $$\sum_{k=0}^\infty {n \choose k}^2 = {2n \choose n}$$ has a combinatorial interpretation: to select $n$ items from $2n$, first take an arbitrary subset of the first $n$ items, and if this had cardinality $k$ select $n-k$ of the second $n$ items.

Similarly,

$$ {2n \choose n} = \sum_{k=1}^{n} {n+1 \choose k} {n-1 \choose n-k}$$

and $$ {n+1 \choose k} {n-1 \choose n-k} = \frac{n+1}{n} {n \choose k}{n \choose k-1} $$

score 2 · Answer 2 · edited Apr 13 '17 at 12:21

Vandermonde's Identity $$ \begin{align} \sum_{k=0}^n\binom{n}{k}^{\large2} &=\sum_{k=0}^n\binom{n}{k}\binom{n}{n-k}\\ &=\binom{2n}{n} \end{align} $$ and $$ \begin{align} \sum_{k=1}^n\binom{n}{k}\binom{n}{k-1} &=\sum_{k=1}^n\binom{n}{k}\binom{n}{n-k+1}\\ &=\binom{2n}{n+1}\\ &=\frac{n}{n+1}\binom{2n}{n} \end{align} $$ which proves the result.

Another Approach Copied From This Answer

Lemma: $$ \sum_{k=1}^n\binom{n}{k}\binom{n}{k-1}=\frac{n}{n+1}\sum_{k=0}^n\binom{n}{k}^2\tag{1} $$ Proof:

Since $\binom{n}{k-1}=\frac{k}{n-k+1}\binom{n}{k}$, we have $\binom{n}{k}+\binom{n}{k-1}=\frac{n+1}{n-k+1}\binom{n}{k}$. Therefore, $$ \frac{n-k+1}{n+1}\left[\binom{n}{k}+\binom{n}{k-1}\right]\binom{n}{k-1}=\binom{n}{k}\binom{n}{k-1}\tag{2} $$ Since $\binom{n}{k}=\frac{n-k+1}{k}\binom{n}{k-1}$, we have $\binom{n}{k}+\binom{n}{k-1}=\frac{n+1}{k}\binom{n}{k-1}$. Therefore, $$ \frac{k}{n+1}\left[\binom{n}{k-1}+\binom{n}{k}\right]\binom{n}{k}=\binom{n}{k-1}\binom{n}{k}\tag{3} $$ Adding $(2)$ and $(3)$ and cancelling yields $$ \frac{n-k+1}{n+1}\binom{n}{k-1}^2+\frac{k}{n+1}\binom{n}{k}^2=\binom{n}{k-1}\binom{n}{k}\tag{4} $$ Summing $(4)$ over $k$, and substituting $k\mapsto k+1$ in the leftmost sum, gives $$ \frac{n}{n+1}\sum_{k=0}^n\binom{n}{k}^2=\sum_{k=1}^n\binom{n}{k-1}\binom{n}{k}\tag{5} $$ QED

My favorite out of the Vandermonde variants. Concise and complete. (+1) — Markus Scheuer, Sep 30 '16 at 09:15

Markus Scheuer · Answer 3 · 2016-09-30T20:00:58.117

Here we show the binomial identity from scratch without using Vandermonde's identity. It is convenient to use the coefficient of operator $[z^k]$ to denote the coefficient of $z^k$ in a series. This way we can write e.g. \begin{align*} [z^k](1+z)^n=\binom{n}{k} \end{align*}

We obtain \begin{align*} \sum_{k=0}^n\binom{n}{k}^2&=\sum_{k=0}^\infty[z^k](1+z)^n[u^k](1+u)^n\tag{1}\\ &=[z^0](1+z)^n\sum_{k=0}^\infty z^{-k}[u^k](1+u)^n\tag{2}\\ &=[z^0](1+z)^n\left(1+\frac{1}{z}\right)^n\tag{3}\\ &=[z^n](1+z)^{2n}\\ &=\binom{2n}{n} \end{align*}

Comment:

In (1) we apply the coefficient of operator twice and set the upper limit of the series to $\infty$ without changing anything since we are adding zeros only.
In (2) we use the linearity of the coefficient of operator and apply the rule $$[z^{p+q}]A(z)=[z^p]z^{-q}A(z)$$
In (3) we use the substitution rule with $u:= \frac{1}{z}$

\begin{align*} A(u)=\sum_{n=0}^\infty a_n u^n=\sum_{n=0}^\infty u^n [z^n]A(z) \end{align*}

We obtain in the same way as above \begin{align*} \frac{n+1}{n}\sum_{k=1}^n\binom{n}{k}\binom{n}{k-1} &=\frac{n+1}{n}\sum_{k=1}^n\frac{n}{k}\binom{n-1}{k-1}\binom{n}{k-1}\\ &=\sum_{k=0}^{n-1}\binom{n-1}{k}\binom{n+1}{k+1}\\ &=\sum_{k=0}^\infty[z^k](1+z)^{n-1}[u^{k+1}](1+u)^{n+1}\\ &=[z^0](1+z)^{n-1}\sum_{k=0}^\infty z^{-k}[u^k]\frac{(1+u)^{n+1}}{u}\\ &=[z^0](1+z)^{n-1}\frac{\left(1+\frac{1}{z}\right)^{n+1}}{\frac{1}{z}}\\ &=[z^n](1+z)^{2n}\\ &=\binom{2n}{n} \end{align*} and the claim follows.

Nice work. (+1). The substitution rule always makes for an interesting effect. — Marko Riedel, Sep 30 '16 at 20:52

Marco Cantarini · Answer 4 · 2016-09-30T07:12:45.010

We can also use the integral representation of the binomial coefficient $$\dbinom{n}{k}=\frac{1}{2\pi i}\oint_{\left|z\right|=1}\frac{\left(1+z\right)^{n}}{z^{k+1}}dz $$ and so $$\begin{align}\frac{n+1}{n}\sum_{k=0}^{n}\dbinom{n}{k}\dbinom{n}{k-1}= & \frac{n+1}{2n\pi i}\oint_{\left|z\right|=1}\left(1+z\right)^{n}\sum_{k=0}^{n}\dbinom{n}{k}z^{-k}dz \\ = & \frac{n+1}{2n\pi i}\oint_{\left|z\right|=1}\left(1+z\right)^{n}\left(1+1/z\right)^{n}dz \\ = & \frac{n+1}{2n\pi i}\oint_{\left|z\right|=1}\frac{\left(1+z\right)^{2n}}{z^{n}}dz \\ = & \frac{n+1}{n}\dbinom{2n}{n-1} \\ = & \color{red}{\dbinom{2n}{n}} \end{align} $$ as wanted, since we have the recurrence $$\frac{m+1-l}{l}\dbinom{m}{l-1}=\dbinom{m}{l}.$$ Using this technique, it is quite simple to show that $$\sum_{k=0}^{n}\dbinom{n}{k}^{2}=\dbinom{2n}{n}.$$

score 0 · Answer 5 · answered Sep 29 '16 at 21:36

I don't have a full answer, but I was thinking maybe this could be useful:

$2^{2n}=\left(\sum_{k=0}^n \binom{n}{k}\right)^2=\sum_{k=0}^n \binom{n}{k}^2+2e_2\left( \binom{n}{j} \right)$ where $e_2\left( x_j\right)$ is the elementary symmetric polynomial on the $n$ symbols $x_j$ for $j=1,...,n$ So that $\sum_{k=0}^n \binom{n}{k}^2=2^{2n}-2e_2\left( \binom{n}{j}\right)$

I will think about this some more to see if it leads anywhere.

This looks more like a comment than an answer to me. – Surb Sep 29 '16 at 21:40 — Surb, Sep 29 '16 at 21:40

score 0 · Answer 6 · answered Sep 30 '16 at 06:10

Apply four equation: $$\sum_{k=0}^\infty\binom{r}{k}\binom{s}{n-k}=\binom{r+s}{n} [1]$$ and $$\binom{n}{k}=\frac{n}{n-k}\binom{n-1}{k} [2]$$ and $$\binom{n}{k}=\binom{n}{n-k} [3]$$ and (for k!= 0) $$\binom{n}{k}=\frac{n}{k}\binom{n-1}{k-1} [4]$$

Apply [1] and [2] we have: $$\sum_{k=0}^{n}\binom{n}{k}^2 = \binom{n+n}{n}=\binom{2n}{n}$$ $$=\frac{2n}{2n-n}\binom{2n-1}{n}=2\binom{2n-1}{n}$$

Note that $\binom{n}{n+1}=0$, Apply [3] and [1] and [4] we have

$$\frac{n+1}{n}\sum_{k=1}^{n}\binom{n}{k}\binom{n}{k-1} = \frac{n+1}{n}\sum_{k=0}^{n+1}\binom{n}{k}\binom{n}{n+1-k}$$ $$ = \frac{n+1}{n}\binom{n+n}{n+1}=\frac{n+1}{n}\binom{2n}{n+1} $$ $$ = \frac{n+1}{n}\frac{2n}{n+1}\binom{2n-1}{n} = 2\binom{2n-1}{n}$$

Conclusion: $$\sum_{k=0}^{n}\binom{n}{k}^2=\frac{n+1}{n}\sum_{k=1}^{n}\binom{n}{k}\binom{n}{k-1} = 2\binom{2n-1}{n}$$

score 0 · Answer 7 · answered Oct 01 '16 at 09:40

Elegant Proof

You have $2n$ students $n$ male, $n$ female. You want to construct a committee of $n$ students and you also want to make one of them president of the committee. You can do this in $$ n{2n \choose n} $$ different ways.

Now say the number of male students are $k$. Then this is also equivalent to choosing $k$ male and $n-k$ female students and making one of them president $$ n \sum \limits_{k=0}^{n-1} {n \choose k} {n \choose n-k} $$
Now lets first choose the non-president $n-1$ committee members then choose the president from the remaining $n+1$ You can do this in $$ (n+1) \sum \limits_{k=0}^{n} {n \choose k} {n \choose n-k+1} $$ different ways. Hence we have

$$ n \sum \limits_{k=0}^{n} {n \choose k} {n \choose n-k} = (n+1) \sum \limits_{k=0}^{n} {n \choose k} {n \choose n-1-k} $$ which is same as $$ n \sum \limits_{k=0}^{n} {n \choose k}^2 = (n+1) \sum \limits_{k=0}^{n} {n \choose k} {n \choose k+1} $$

Felix Marin · Answer 8 · 2016-10-04T06:36:49.763

$\newcommand{\braces}[1]{\left\lbrace\,{#1}\,\right\rbrace} \newcommand{\bracks}[1]{\left\lbrack\,{#1}\,\right\rbrack} \newcommand{\dd}{\mathrm{d}} \newcommand{\ds}[1]{\displaystyle{#1}} \newcommand{\expo}[1]{\,\mathrm{e}^{#1}\,} \newcommand{\ic}{\mathrm{i}} \newcommand{\mc}[1]{\mathcal{#1}} \newcommand{\mrm}[1]{\mathrm{#1}} \newcommand{\pars}[1]{\left(\,{#1}\,\right)} \newcommand{\partiald}[3][]{\frac{\partial^{#1} #2}{\partial #3^{#1}}} \newcommand{\root}[2][]{\,\sqrt[#1]{\,{#2}\,}\,} \newcommand{\totald}[3][]{\frac{\mathrm{d}^{#1} #2}{\mathrm{d} #3^{#1}}} \newcommand{\verts}[1]{\left\vert\,{#1}\,\right\vert}$

$\ds{\sum_{k = 0}^{n}{n \choose k}^{2} = {n + 1 \over n}\sum_{k = 1}^{n}{n \choose k}{n \choose k - 1}:\ ?}$.

\begin{align} &\color{#f00}{{n + 1 \over n}\sum_{k = 1}^{n}{n \choose k}{n \choose k - 1}} = {n + 1 \over 2n}\sum_{k = 1}^{n}\braces{% \bracks{{n \choose k} + {n \choose k - 1}}^{2} - {n \choose k}^{2} - {n \choose k - 1}^{2}} \\[5mm] = &\ {n + 1 \over 2n}\sum_{k = 1}^{n}{n + 1 \choose k}^{2} - {n + 1 \over 2n}\sum_{k = 1}^{n}{n \choose k}^{2} - {n + 1 \over 2n}\sum_{k = 1}^{n}{n \choose k - 1}^{2} \quad\pars{\begin{array}{l}Pascal\ Triangle\ Identity\ \\ \mbox{in the first term}\end{array}} \\[5mm] = &\ {n + 1 \over 2n}\bracks{-1 + \sum_{k = 0}^{n + 1}{n + 1 \choose k}^{2} - 1} - {n + 1 \over 2n}\bracks{-1 + \sum_{k = 0}^{n}{n \choose k}^{2}} - {n + 1 \over 2n}\bracks{\sum_{k = 0}^{n}{n \choose k}^{2} - 1} \\[5mm] = &\ \color{#f00}{{n + 1 \over 2n}\sum_{k = 0}^{n + 1}{n + 1 \choose k}^{2} - {n + 1 \over n}\sum_{k = 0}^{n}{n \choose k}^{2}}\label{1}\tag{1} \end{align}

\begin{equation} \mbox{We'll use the}\ well\ known\ identity\quad \sum_{j = 0}^{m}{m \choose j}^{2} = {2m \choose m}\label{2}\tag{2} \end{equation}

Expression \eqref{1} becomes: \begin{align} &\color{#f00}{{n + 1 \over n}\sum_{k = 1}^{n}{n \choose k}{n \choose k - 1}} = {n + 1 \over 2n}{2n + 2 \choose n + 1} - {n + 1 \over n}\sum_{k = 0}^{n}{n \choose k}^{2} \\[5mm] = &\ {n + 1 \over 2n}\bracks{{\pars{2n + 2}\pars{2n + 1} \over \pars{n + 1}\pars{n + 1}}{2n \choose n}} - {n + 1 \over n}\sum_{k = 0}^{n}{n \choose k}^{2} = {2n + 1 \over n}\ \overbrace{\sum_{k = 0}^{n}{n \choose k}^{2}} ^{\ds{\mbox{see}\ \eqref{2}}}\ -\ {n + 1 \over n}\sum_{k = 0}^{n}{n \choose k}^{2} \\[5mm] = &\ \color{#f00}{\sum_{k = 0}^{n}{n \choose k}^{2}} \end{align}

Show that $\sum_{k=0}^{n}\binom{n}{k}^2=\frac{n+1}{n}\sum_{k=1}^{n}\binom{n}{k}\binom{n}{k-1}$

8 Answers8