Bob's nested sum

For positive integers m m and n n , let

f ( m , n ) = a 1 = 1 m ( a 2 = 1 a 1 ( a 3 = 1 a 2 ( ( a n 1 = 1 a n 2 ( a n = 1 a n 1 a n ) ) ) ) ) . f(m,n) = \displaystyle \sum_{a_1=1}^{m} \left(\displaystyle \sum_{a_2=1}^{a_1} \left(\displaystyle \sum_{a_3=1}^{a_2} \left(\cdots \left(\displaystyle \sum_{a_{n-1}=1}^{a_{n-2}} \left(\displaystyle \sum_{a_n=1}^{a_{n-1}} a_n\right)\right) \cdots \right)\right)\right).

How many values of n 1000 n\leq 1000 satisfy f ( 11 , n ) f ( 12 , n ) f(11,n) \mid f(12,n) ?

This problem is posed by Bob K .

Details and assumptions

For n = 1 n=1 , f ( m , n ) = a 1 = 1 m a 1 f(m,n) = \sum_{a_1=1}^m a_1 .


The answer is 91.

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.

4 solutions

Johan de Ruiter
Sep 11, 2013

We can replace the summand a n a_n with a n + 1 = 1 a n 1 \sum_{a_{n+1}=1}^{a_n} 1 . We are now counting the number of (reversed) sorted vectors of length n + 1 n+1 with maximum value m m . If we add m 1 m - 1 dividers between n + 1 n+1 items, we can indicate how many of each value in [ 1 , m ] [1, m] we pick. There are ( n + 1 + m 1 m 1 ) = ( n + m m 1 ) {n + 1 + m - 1 \choose m - 1} = {n + m \choose m - 1} ways to do this.

So when does ( n + 11 10 ) {n + 11 \choose 10} divide ( n + 12 11 ) {n + 12 \choose 11} ?

( n + 11 ) ! 10 ! × ( n + 1 ) ! ( n + 12 ) ! 11 ! × ( n + 1 ) ! \frac{(n+11)!}{10! \times (n+1)!} \vert \frac{(n+12)!}{11! \times (n+1)!}

11 ( n + 12 ) 11 \vert (n + 12)

11 ( n + 1 ) 11 \vert (n + 1)

Since 11 11 divides 1001 1001 , the answer is 1001 11 = 91 \frac{1001}{11} = 91 .

Yup! Great simple solution. :D

Finn Hulse - 7 years ago
Mark Hennings
Sep 9, 2013

We have f ( m , 1 ) = r = 1 m r = 1 2 m ( m + 1 ) = ( m + 1 2 ) f(m,1) \; = \; \sum_{r=1}^m r \; = \; \tfrac{1}{2}m(m+1) \; = \; {m+1 \choose 2} If n 1 n \ge 1 and f ( m , n ) = ( m + n n + 1 ) f(m,n) \; = \; {m+n \choose n+1} then f ( m , n + 1 ) = r = 1 m f ( r , n ) = r = 1 m ( r + n n + 1 ) = ( m + n + 1 n + 2 ) f(m,n+1) \; = \; \sum_{r=1}^m f(r,n) \; = \; \sum_{r=1}^m {r+n \choose n+1} \; =\; {m+n+1 \choose n+2} Hence, by induction on n n , f ( m , n ) = ( m + n n + 1 ) f(m,n) \,=\, {m+n \choose n+1} , and hence f ( 12 , n ) f ( 11 , n ) = n + 12 11 \frac{f(12,n)}{f(11,n)} \; = \; \frac{n+12}{11} Thus we want to know the number of n n between 1 1 and 1000 1000 such that 11 11 divides n + 12 n+12 , namely such that n 10 n \equiv 10 modulo 11 11 . Possible solutions are 10 , 21 , , 1000 = 10 + 90 × 11 10,21,\ldots,1000=10+90\times11 , and so there are 91 91 values of n n .

D G
Sep 8, 2013

f(m, n) is equivalent to the binomial coefficient of (n + m, m - 1). Therefore:

f ( 11 , n ) = b i n ( 11 + n , 10 ) = ( 11 + n ) ! 10 ! ( n + 1 ) ! f(11, n) = bin(11 + n, 10) = \frac{(11 + n)!}{10! * (n + 1)!} f ( 12 , n ) = b i n ( 12 + n , 11 ) = ( 12 + n ) ! 11 ! ( n + 1 ) ! f(12, n) = bin(12 + n, 11) = \frac{(12 + n)!}{11! * (n + 1)!}

"|" denotes "is divisible by".

Dividing these two equations and applying some simple algebra:

f ( 12 , n ) / f ( 11 , n ) = n + 12 11 f(12, n) / f(11, n) = \frac{n + 12}{11}

Clearly this only holds when n = = 10 ( m o d 11 ) n == 10 (mod 11) , and there are 91 such values below 1000.

Matt McNabb
Sep 13, 2013

Examining the definition for f f , we observe that if n > 1 n>1 , then f ( m + 1 , n ) = f ( m , n ) + f ( m + 1 , n 1 ) f(m+1,n) = f(m,n) + f(m+1,n-1)

Also we find that f ( 1 , n ) = 1 f(1,n) = 1 f ( m , 1 ) = ( 1 + 2 + . . . + m ) f(m,1)=(1+2+...+m)

Now, I recognized this as the iterative rule and the boundary conditions for Pascal's Triangle: each element is the sum of the two elements above it. The general formula for an element of this triangle is f ( m , n ) = ( m + n m 1 ) f(m,n) = {m+n \choose m-1}

To prove this, we can inspect that it fulfils the initial conditions correctly. Also, the iterative step f ( m , n ) + f ( m + 1 , n 1 ) = ( m + n m 1 ) + ( m + n m ) = ( m + n + 1 m ) = f ( m + 1 , n ) \begin{aligned} f(m,n) + f(m+1,n-1) &= {m+n \choose m-1} + {m+n \choose m} \\ &= {m+n+1 \choose m} \\ &= f(m+1,n) \end{aligned}

Moving on, we have f ( 12 , n ) = ( 12 + n 11 ) = ( 12 + n ) ! 11 ! ( n + 1 ) ! = ( 11 + n ) ! 10 ! ( n + 1 ) 12 + n 11 = f ( 11 , n ) 12 + n 11 \begin{aligned} f(12,n) &= {12+n \choose 11} \\ &= { (12+n)! \over 11!(n+1)!} \\ &= { (11+n)! \over 10!(n+1)} {12+n \over 11} \\ &= f(11,n) {12+n \over 11} \end{aligned}

So we simply have to find how many values of n n satisfy 11 n + 12 11 \mid n+12 .

n n can range from 10 = 11 1 1 10 = 11*1-1 to 1000 = 11 91 1 1000 = 11*91-1 so there are 91 \boxed{91} values.

0 pending reports

×

Problem Loading...

Note Loading...

Set Loading...