Distributive factorials

How many ordered pairs of positive integers 1 k n 50 1 \leq k \leq n \leq 50 are there, such that k k divides n n , and

( n k ) ! = n ! k ! ? \left(\frac{n}{k}\right)! = \frac{n!}{k!} \quad ?

Details and assumptions

For an ordered pair of integers ( a , b ) (a,b) , the order of the integers matter. The ordered pair ( 1 , 2 ) (1, 2) is different from the ordered pair ( 2 , 1 ) (2,1) .


The answer is 99.

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.

10 solutions

Calvin Lin Staff
May 13, 2014

Solution 1: When k = 1 k = 1 this is clearly satisfied for all n n , since the equation simplifies to n ! = n ! n! = n! . When n = k n = k , the equation is again satisfied. The right hand side will be a product of n k n - k consecutive integers, the largest of which is n n . The left hand side will be a product of n k \frac{n}{k} consecutive integers, the largest of which is n k \frac{n}{k} . When n k > n k n - k > \frac{n}{k} , this equation can never be equal. This inequality can be rearranged as n > k + 1 + 1 k 1 , k 1 n >k + 1 + \frac{1}{k-1}, k \neq 1 . Since k n k \mid n , n k n \geq k , this leaves n = k + 1 n = k+1 and k = 2 , n = 4 k = 2, n = 4 as the only possibilities not excluded by this, or already considered. When n = k + 1 n = k + 1 and k n k \mid n , we have k k + 1 k \mid k + 1 , so k 1 k \mid 1 and thus k = 1 k = 1 . We have already considered this case. When k = 2 , n = 4 k = 2, n = 4 , we have 2 = ( 4 2 ) ! 4 ! 2 ! = 12 2 = \left(\frac{4}{2}\right)! \neq \frac{4!}{2!} = 12 .

Thus we see that the only possible solutions are when k = 1 k = 1 or n = k n = k , by the principle of inclusion and exclusion, there are 50 + 50 1 = 99 50 + 50 - 1 = 99 such solutions.

Solution 2: Here is another way to justify that k k must be 1 1 or n n . Suppose n = k m . n=km. Then
k ! m ! = ( 1 2 . . . k ) ( 1 2 . . . m ) = 1 k m 1 ( 1 . . . ( k 1 ) ) ( ( k 1 ) . . . ( k m ) ) 1 k m 1 ( k m ) ! k!m!=(1\cdot 2 \cdot ...\cdot k)\cdot (1\cdot 2 \cdot...\cdot m ) =\frac{1}{k^{m-1}}\left(1\cdot ...\cdot(k-1)\right)\cdot \left((k\cdot 1)\cdot ...\cdot(k\cdot m)\right)\leq \frac{1}{k^{m-1}}(km)! So if both k k and m m are greater than 1 1 , k ! m ! < ( k m ) ! k!m!<(km)!

Ah, thank you! Had to scroll through way too many incomplete proofs to get to one nice one.

Ryan Tamburrino - 6 years, 6 months ago

Loved solution 2. < 3 <3

Soumava Pal - 5 years, 3 months ago
Md. Imrul Hassan
May 20, 2014

The equation can only be true if either n=k or k=1. There are 50 pairs of values where n=k and 50 pairs of values where k=1. However, n=1, k=1 satisfies both n=k and k=1 and as such counted twice. So, using inclusion-exclusion principle total number of pairs = 50+50-1=99

This solutions is incomplete, because it is not proven that k must be either n or 1. This was a very common mistake. Indeed it is much easier to find a correct answer in this problem, than to justify it.

Calvin Lin Staff - 7 years ago
Shivang Agarwal
May 20, 2014

for k=1,n can take 50 values from 1 to 50 which will satisfy our condition so 50 ordered pairs ; and 49 pairs of n=k such as(2,2),(3,3)........................(50,50) [not including(1,1) because it is included in the above case]

therefore total ordered pairs=99

It is not justified that there are no other solutions

Calvin Lin Staff - 7 years ago
Bruno Moderno
May 20, 2014

Let x=n/k, x integer. We need an x such as x! = n!/k!. So, n!=k!x!. But n=kx. Then, we need an x and a k such as (kx)!=k!x!. That only happens when x=1 or k=1. If k=1, n can be from 1 to 50. If x=1, n=k and n can be from 1 to 50. We have 50 pairs on the first one and 50 on the second. But the solution (1,1) is counted two times. So, we there are 99 solutions.

"That only happens when x=1 or k=1."

Calvin Lin Staff - 7 years ago
Akki Chaudhary
May 20, 2014

as (n/k)! would be too small in comparison to n!/k! for distinct n and k.either k=1 or k=n can satisty the given relation.

"as (n/k)! would be too small" basically correct, but not justified

Calvin Lin Staff - 7 years ago
Otávio Sales
May 20, 2014

Para que a igualdade seja satisfeita há apenas duas condições:

1) Que n=k; ou 2) Que k=1.

São 50 números, de 1 até 50, cada um deles apresenta-se em 2 das duas condições, com exceção do 1, onde ambas condições são idênticas. (Todo número divide por 1 e por ele mesmo, 1 é ele mesmo)

Portanto 2x50-1=99.

1) A igualdade é satisfeita se n=k, por motivos óbvios. (n/k)!=(n/n)!=1!=1 (1o membro) n!/k!=n!/n!=1 (2o membro)

2) A igualdade também é satisfeita se k=1, também por motivos óbvios: (n/1)!=n! n!/1!=n!/1=n

É fácil ver que em quaisquer outras condições a igualdade não é satisfeita: Se k for um divisor de n, (n/k)! é um fatorial de um certo número. Já n!/k! será o número n(n-1)....(k+1).1 essa expressão só é um fatorial se k+1=2 (colocamos o 1 propositalmente para compreensão da idéia) ou se n=k (basta ler o produto 'ao contrário')

Judging by the Google translation, the fact that k=1 or n is not justified.

Calvin Lin Staff - 7 years ago
Arnav Sastry
May 20, 2014

The above holds true iff (k=1) || (n==k). The union of those sets has size 99 (n=1 and k=1 is counted twice). We can see this because no factorial can equal a greater factorial divided by a lesser factorial.

"The above holds true iff (k=1) || (n==k)." The "only if" part is not totally obvious.

Calvin Lin Staff - 7 years ago

let n =kt => k!t!=(kt)! 1 2 3 4 ... k ((LHS.))= (t+1)(t+2)(t+3)...(kt) ((RHS)) >=2 3 4 ...*k

LHS.is multiple of the number k RHS.is multiple of the number kt-t So k >= kt-t <=> k+t >= kt

case 1) k > t ; k+k > k+t >= kt so 2 > t => t =1 is posible (and work!)

case 2) k < t ; so k = 1 (and work!)

case 3) k = t ; if t >2 Apparently LHS. < RHS. so k = t = 1 (and work! lol)

so (n,k)=(kt,k) is possible is n=k or k = 1 have 50 + 50 -1 =99 #

Jorge Fernández
Jul 16, 2015

k must be 1 or n, proof by contradiction(suppose k is not 2 or n: (n/k)! is n! divided by the product of all integers between n and (n/k)+1. There are n-n/k such integers. Notice n/k is between 2 and n/2-1, so 1>n-n/k>=k. So we have a product of at least k integers, each larger than k. Hence this product is larger than k!.

From here (n/k)!<n!/k!

Abhishek De
May 20, 2014

By Bertrand's Postulate, \exists prime p [ n , 2 n ] p\in [n,2n] . Since n ! = ( n k ) ! k ! n!=\left(\frac{n}{k}\right)!k! we have p k p|k or p n k p|\frac{n}{k} which means either k > n 2 k>\frac{n}{2} or n k > n 2 k = 1 , n \frac{n}{k}>\frac{n}{2}\implies k=1,n Thus n = 1 k = 1 n=1\implies k=1

n = 2 k = 1 , 2 n=2\implies k=1,2

.

.

.

n = 50 k = 1 , 50 n=50\implies k=1,50

Thus 99 99 ordered pairs in total.

"By Bertrand's Postulate, \exists prime p [ n , 2 n ] p\in [n,2n] . " True, but totally irrelevant

Calvin Lin Staff - 7 years ago

0 pending reports

×

Problem Loading...

Note Loading...

Set Loading...