Think you can solve this?

k = 0 n [ ( 1 ) k ( n k ) ( n k ) n ] = ? \large \sum_{k=0}^n \left[ (-1)^k \dbinom{n}{k} (n-k)^n \right] = \ ?

n ! n! ( n k ) ! (n-k)! ( n 1 ) ! (n-1)! 0 0 1 1

This section requires Javascript.
You are seeing this because something didn't load right. We suggest you, (a) try refreshing the page, (b) enabling javascript if it is disabled on your browser and, finally, (c) loading the non-javascript version of this page . We're sorry about the hassle.

7 solutions

Nelson Mandela
Aug 16, 2015

I will try to use functions to prove this.

A bijective is defined as a function which is both one-one and onto(surjective).

In a bijective, number of elements in domain = number of elements of co-domain = (n) and range is equal to co-domain.

In a bijective, No. of one-one functions is equal to number of onto functions.

No.of. onto functions = k = 0 n ( 1 ) k ( n k ) ( n k ) m \sum_{k=0}^n(-1)^k\binom{n}k(n-k)^m .

Here, m = n.

So, k = 0 n ( 1 ) k ( n k ) ( n k ) n \sum_{k=0}^n(-1)^k\binom{n}k(n-k)^n =no.of.one-one functions from n to n = n!.

Also, I think the power of -1 should be k not n.

Please correct me if I am wrong.

yes you are absolutely correct! yes it should be k im sorry

Gokul Kumar - 5 years, 10 months ago

why is the number of onto functions the sum?

The Physicist Cuber Mauro - 3 years, 4 months ago

How did you get to the conclusion that its is a bijective function?

Mohit Marathe - 1 year, 11 months ago
Alan Yan
Jun 25, 2017

Suppose that we have n n distinct stones and n n distinct colors. We will calculate the number of ways to paint our stones such that each color is used. Obviously, this number is n ! n! . But, from PIE and complementary counting (we consider the cases where we don't use a finite number of colors), the number of ways is also the expression in the hypothesis. Thus, since they count the same quantity, the expressions must be equal. Hence, the answer is n ! n! .

Could you elaborate on how PIE and complementary counting gives you the same expression as the one in the problem?

Ben Lou - 3 years, 3 months ago
Zhizhong Lin
Jun 7, 2020

I will use inductions to prove this.

Lemma 1: k = 0 n ( 1 ) k ( n k ) ( n k ) = 0 \sum_{k=0}^{n}(-1)^{k}\binom{n}{k}(n-k)=0 for all n 2 n \ge 2

Proof:

k = 0 n ( 1 ) k ( n k ) ( n k ) = n k = 0 n ( 1 ) k ( n k ) k = 1 n ( 1 ) k ( n k ) k \sum_{k=0}^{n}(-1)^{k}\binom{n}{k}(n-k)=n\sum_{k=0}^{n}(-1)^{k}\binom{n}{k}-\sum_{k=1}^{n}(-1)^{k}\binom{n}{k}k

Since ( n k ) k = ( n 1 k 1 ) n \binom{n}{k}k=\binom{n-1}{k-1}n and 0 = ( 1 1 ) n = k = 0 n ( 1 ) k ( n k ) 0=(1-1)^n=\sum_{k=0}^{n}(-1)^{k}\binom{n}{k} , we have

k = 0 n ( 1 ) k ( n k ) ( n k ) = n k = 1 n ( 1 ) k ( n 1 k 1 ) = n i = 0 n 1 ( 1 ) i ( n 1 i ) = n ( 1 1 ) n 1 = 0 \sum_{k=0}^{n}(-1)^{k}\binom{n}{k}(n-k)=-n\sum_{k=1}^{n}(-1)^{k}\binom{n-1}{k-1}=n\sum_{i=0}^{n-1}(-1)^{i}\binom{n-1}{i}=n(1-1)^{n-1}=0

The proof is done.

Lemma 2: k = 0 n ( 1 ) k ( n k ) ( n k ) m = 0 \sum_{k=0}^{n}(-1)^{k}\binom{n}{k}(n-k)^{m} = 0 for all n m + 1 n \ge m+1

Proof: (Induction)

By Lemma 1 , Lemma 2 holds when m = 1. Suppose k = 0 n ( 1 ) k ( n k ) ( n k ) m = 0 \sum_{k=0}^{n}(-1)^{k}\binom{n}{k}(n-k)^{m} = 0 holds for n m + 1 n\ge m+1 , then for n m + 2 n\ge m+2 ,

k = 0 n ( 1 ) k ( n k ) ( n k ) m + 1 = n k = 0 n ( 1 ) k ( n k ) ( n k ) m k = 0 n ( 1 ) k ( n k ) k ( n k ) m \sum_{k=0}^{n}(-1)^{k}\binom{n}{k}(n-k)^{m+1}=n\sum_{k=0}^{n}(-1)^{k}\binom{n}{k}(n-k)^{m}-\sum_{k=0}^{n}(-1)^{k}\binom{n}{k}k(n-k)^{m}

The first term on the RHS equals 0 by our assumption since n m + 2 m + 1 n\ge m+2\ge m+1 , and the second term equals n k = 1 n ( 1 ) k ( n 1 k 1 ) ( n k ) m = n k = 1 n ( 1 ) k 1 ( n 1 k 1 ) ( n k ) m = n i = 0 n 1 ( 1 ) i ( n 1 i ) ( n 1 i ) m -n\sum_{k=1}^{n}(-1)^{k}\binom{n-1}{k-1}(n-k)^{m}=n\sum_{k=1}^{n}(-1)^{k-1}\binom{n-1}{k-1}(n-k)^{m}=n\sum_{i=0}^{n-1}(-1)^{i}\binom{n-1}{i}(n-1-i)^{m} . This term equals 0 by our assumption since n 1 m + 1 n-1\ge m+1 . So

k = 0 n ( 1 ) k ( n k ) ( n k ) m + 1 = 0 \sum_{k=0}^{n}(-1)^{k}\binom{n}{k}(n-k)^{m+1}=0 for all n m + 2 n\ge m+2

The proof is done.

Corollary: k = 0 n ( 1 ) k ( n k ) ( n k ) n 1 = 0 \sum_{k=0}^{n}(-1)^{k}\binom{n}{k}(n-k)^{n-1} = 0

Proof: This is a special case of Lemma 2 when m=n-1

The proof is done.

Theorem: k = 0 n ( 1 ) k ( n k ) ( n k ) n = n ! \sum_{k=0}^{n}(-1)^{k}\binom{n}{k}(n-k)^{n} = n!

Proof: The equation holds when n=1. Suppose k = 0 m ( 1 ) k ( m k ) ( n k ) m = m ! \sum_{k=0}^{m}(-1)^{k}\binom{m}{k}(n-k)^{m} = m! holds when m n m\le n , where n 1 n\ge 1 , then

k = 0 m + 1 ( 1 ) k ( m + 1 k ) ( m + 1 k ) m + 1 = ( m + 1 ) k = 0 m + 1 ( 1 ) k ( m + 1 k ) ( m + 1 k ) m k = 0 m + 1 ( 1 ) k ( m + 1 k ) k ( m + 1 k ) m \sum_{k=0}^{m+1}(-1)^{k}\binom{m+1}{k}(m+1-k)^{m+1}=(m+1)\sum_{k=0}^{m+1}(-1)^{k}\binom{m+1}{k}(m+1-k)^{m}-\sum_{k=0}^{m+1}(-1)^{k}\binom{m+1}{k}k(m+1-k)^{m}

The first term on the RHS equals 0 by the Corollary , and the second term

k = 0 m + 1 ( 1 ) k ( m + 1 k ) k ( m + 1 k ) m = ( m + 1 ) k = 1 m + 1 ( 1 ) k 1 ( m k 1 ) ( m ( k 1 ) ) m = ( m + 1 ) i = 0 m ( 1 ) i ( m i ) ( m i ) m = ( m + 1 ) m ! = ( m + 1 ) ! -\sum_{k=0}^{m+1}(-1)^{k}\binom{m+1}{k}k(m+1-k)^{m}=(m+1)\sum_{k=1}^{m+1}(-1)^{k-1}\binom{m}{k-1}(m-(k-1))^{m}=(m+1)\sum_{i=0}^{m}(-1)^{i}\binom{m}{i}(m-i)^{m} = (m+1)m! = (m+1)!

So k = 0 m + 1 ( 1 ) k ( m + 1 k ) ( m + 1 k ) m + 1 = ( m + 1 ) ! \sum_{k=0}^{m+1}(-1)^{k}\binom{m+1}{k}(m+1-k)^{m+1} = (m+1)!

The proof is done.

Aditya Gupta
Nov 21, 2018

Ok so clearly you all guys are too intelligent to come up with such short and elegant solutions involving extreme logical thinking rather than pure algebraic maneuvers. But for a slow person like me who finds it toooo difficult to relate a summation to the number of ways of counting an event, I had to spend half an hour solving it explicitly by using formulas and identities: Write S= summation (-1)^(n-k).nCk.k^n (coz f(k) is replaced by f(a+b-k)) Now simply write k^n= k(k-1)...(k-(n-1))+ak(k-1)...(k-(n-2))+bk(k-1)...(k-(n-3))+.........+hk, where a, b,.....,h are undetermined coefficients. Using r.nCr= n.(n-1)C(r-1) repeatedly we will see that all these sums vanish except for the first term k(k-1)....(k-(n-1)).(-1)^(n-k).nCk. So we simply take its sum which is precisely it's value at n, i.e., n!

Clarence Zhuo
Feb 17, 2020

The result is trivial if you know the Stirling number of the second kind: S ( m , n ) = 1 n ! k = 0 n ( 1 ) k ( n k ) ( n k ) m . S(m, n) = \frac{1}{n!} \sum_{k=0}^n (-1)^k {n \choose k} (n-k)^m. Let m = n m = n and we get S ( n , n ) = 1 S(n,n) = 1 .

Alon Shapira
Jan 13, 2019

Here are the details for Nelson's solution (same as Alan's solution). I am writing it here, rather than as a reply, because of the latex support.

Let A , B A,B be finite sets.

For a subset B B B'\subset B let E ( B ) E(B') be the set of functions A ( B B ) A\rightarrow (B\setminus B') and for b B b\in B let's write E ( b ) E(b) instead of E ( { b } ) E(\{b\}) for short. Clearly E ( B ) = b B E ( b ) E(B') = \cap_{b \in B'} E(b) and E ( B ) = ( B B ) A |E(B')| = (|B| - |B'|) ^{|A|}

Clearly, b B E ( b ) \cup_{b \in B'} E(b) is the set of all functions which are not onto, and therefore by PIE : onto functions = B B ( 1 ) B E ( B ) = k = 0 B ( 1 ) k ( B k ) ( B k ) A |\text{onto functions} | = \sum_{B'\subset B} (-1)^{|B'|} |E(B')| = \sum_{k=0}^{|B|} (-1)^k \cdot {|B|\choose k} \cdot (|B|-k)^{|A|} Taking A = B = n |A| = |B| = n the identity follows (because there are n ! n! onto functions in this case).

It also follows that m < n k = 0 n ( 1 ) k ( n k ) ( n k ) m = 0 m < n \ \Longrightarrow \sum_{k=0}^{n} (-1)^k \cdot {n\choose k} \cdot (n-k)^{m} =0

Gokul Kumar
Aug 16, 2015

My solution is similar to Nelson's solution. I want to know if anyone has proved this equation using any other method. please do post if you have proved it by some other method!

Yes, I would like to know any other way to solve this.

Nelson Mandela - 5 years, 9 months ago

Log in to reply

Here is another solution...

( x 1 ) n = k = 0 n ( 1 ) k ( n k ) x ( n k ) (x-1)^n = \displaystyle \sum_{k=0}^n (-1)^k {n \choose k} x^{(n-k)}

Now differentiate wrt x on both sides...

n ( x 1 ) n 1 = k = 0 n ( 1 ) k ( n k ) ( n k ) x ( n k 1 ) n(x-1)^{n-1} = \displaystyle \sum_{k=0}^n (-1)^k {n \choose k} (n-k) x^{(n-k-1)}

Multiply both sides by x and then differentiate again...

n x ( x 1 ) n 1 = k = 0 n ( 1 ) k ( n k ) ( n k ) x ( n k ) nx(x-1)^{n-1} = \displaystyle \sum_{k=0}^n (-1)^k {n \choose k} (n-k) x^{(n-k)}

n ( n 1 ) x ( x 1 ) n 2 + n ( x 1 ) n 1 = k = 0 n ( 1 ) k ( n k ) ( n k ) 2 x ( n k 1 ) n(n-1)x(x-1)^{n-2} + n(x-1)^{n-1} = \displaystyle \sum_{k=0}^n (-1)^k {n \choose k} (n-k)^2 x^{(n-k-1)}

If we continue this for n steps...

n ! + ( x 1 ) L ( x ) = k = 0 n ( 1 ) k ( n k ) ( n k ) n x ( n k 1 ) n! + (x-1) L(x) = \displaystyle \sum_{k=0}^n (-1)^k {n \choose k} (n-k)^n x^{(n-k-1)}

where L(x) is some sophisticated polynomial in x. Now put x = 1,

n ! = k = 0 n ( 1 ) k ( n k ) ( n k ) n n! = \displaystyle \sum_{k=0}^n (-1)^k {n \choose k} (n-k)^n

Nishant Shelar - 5 years, 3 months ago

Log in to reply

Nice solution... :)

M D - 4 years, 9 months ago

I used that same method to solve the problem. It wad easy

Pranav Jayaprakasan UT - 4 years, 9 months ago

Induction. Case is clearly true when n=1. Assume n is true, easy to prove n+1 true by expanding n choose k in factorial form and using algebraic manipulation with the (n-k) and also observing that last term is aways zero.

Lau Eng Choon - 3 years, 7 months ago

improve more

improve more - 2 years, 9 months ago

0 pending reports

×

Problem Loading...

Note Loading...

Set Loading...