One... Two... Three... Infinity!

Algebra Level 2

The sequence 1 , 2 , 3 , 2 , 1, 2, 3, 2, \ldots has the property that every fourth term is the average of the previous three terms. That is, the sequence { a n } \{a_n\} is defined as a 1 = 1 , a 2 = 2 , a 3 = 3 a_1 = 1, a_2 = 2, a_3 = 3 , and a n + 3 = a n + 2 + a n + 1 + a n 3 a_{n+3} = \dfrac{a_{n+2} + a_{n+1} + a_n}{3} for any positive integer n n .

This sequence converges to a rational number a b \frac{a}{b} , where a a and b b are coprime positive integers. Find a + b a+b .


The answer is 10.

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

Let a = lim a n a =\lim a_n and x n = 3 a n + 2 + 2 a n + 1 + a n x_n = 3a_{n+2} + 2a_{n+1} + a_n So, x n + 1 = x n x_{n+1} = x_n . Indeed, 3 a n + 3 = a n + 2 + a n + 1 + a n 3a_{n+3} = a_{n+2} + a_{n+1} + a_n \Leftrightarrow 3 a n + 3 + 2 a n + 2 + a n + 1 = 3 a n + 2 + 2 a n + 1 + a n 3a_{n+3} + 2a_{n+2} + a_{n+1} = 3a_{n+2} + 2a_{n+1} + a_n \Leftrightarrow x n + 1 = x n x_{n+1}=x_n Then, x n = x 1 = 3 2 + 2 2 + 1 2 = 14 x_n = x_1 = 3^2 + 2^2+1^2= 14 . In other hand, we have lim x n = lim 3 a n + 2 + 2 a n + 1 + a n = 3 a + 2 a + a = 6 a \lim x_n = \lim 3a_{n+2} + 2a_{n+1} + a_n = 3a + 2a + a = 6a \Rightarrow a = lim x n 6 = x 1 6 = 7 3 a = \dfrac{\lim x_n}{6} = \dfrac{x_1}{6} = \dfrac{7}{3}

How did you get to the idea of writing xn in this manner?

Ariijit Dey - 3 years, 9 months ago

This logic also works for one of the previous problems and would be a better solution than the top one.

Mr Master - 2 years, 6 months ago

How do you prove the convergence of the series?

Maxence Seymat - 2 years, 1 month ago

Log in to reply

it's given

ilayo8 . - 9 months, 4 weeks ago

Log in to reply

Thanks for your reply, but it doesn't answer my question. I don't stop at what's required, I want to understand in-depth.

Maxence Seymat - 9 months ago
Rajen Kapur
Jan 13, 2015

3 a 4 = a 1 + a 2 + a 3 3a_4 = a_1 + a_2 + a_3 3 a 5 = a 2 + a 3 + a 4 3a_5 = a_2 + a_3 + a_4 3 a 6 = a 3 + a 4 + a 5 3a_6 = a_3 + a_4 + a_5 3 a 7 = a 4 + a 5 + a 6 3a_7 = a_4 + a_5 + a_6 and so on and on finally converging to 3 a @ = a @ + a @ + a @ 3a_@ = a_@ + a_@ + a_@ 3 a @ = a @ + a @ + a @ 3a_@ = a_@ + a_@ + a_@ The subscript @ has been used to show the term at infinity (converged). Now adding all these equations three a 4 s a_4 s in L.H.S. cancel with three a 4 s a_4 s in R.H.S. and so do all others. We finally get 6 a @ = a 1 + 2 a 2 + 3 a 3 6a_@ = a_1 + 2a_2 + 3a_3 Now it is easy to see that the rational number is (1 + 2x2 + 3x3)/6 = 7/3. Hence a = 7, b = 3. a + b = 10

You should try the problem It's trivial?It is the same as this problem.

Akshay Bodhare - 6 years, 5 months ago
Roy Tu
Jan 16, 2015

Weird solution:

The recursive operation is given by the matrix:

A = ( 1 3 1 3 1 3 1 0 0 0 1 0 ) A=\left(\begin{array}{c c c} \frac{1}{3} & \frac{1}{3} & \frac{1}{3} \\ 1 & 0 & 0 \\ 0 & 1 & 0 \\ \end{array}\right)

We want to find:

lim n A n ( 3 2 1 ) \lim_{n\rightarrow\infty}A^n\left(\begin{array}{c}3 \\ 2 \\ 1 \\ \end{array}\right)

which we can do by diagonalizing A. After some pain you end up with: ( 1 2 1 3 1 6 ? ? ? ? ? ? ) ( 3 2 1 ) = ( 7 3 7 3 7 3 ) \left(\begin{array}{c c c} \frac{1}{2} & \frac{1}{3} & \frac{1}{6} \\ ? & ? & ? \\ ? & ? & ? \\ \end{array}\right) \left(\begin{array}{c} 3 \\ 2 \\ 1 \\ \end{array}\right)= \left(\begin{array}{c} \frac{7}{3} \\ \frac{7}{3} \\ \frac{7}{3} \\ \end{array}\right)

where we don't care what the ?'s are because we know that as the sequence converges our column vector elements become equivalent. Hence our final result is 7 3 \frac{7}{3} and a + b = 10 a+b=\boxed{10}

Can someone please link me to some study text about what is going on in this solution (I am familiar with basic matrix stuff) :) Thank you

Simona Vesela - 6 years, 4 months ago

Log in to reply

Search eigen-value and eigen-vectors and diagonalization of matrix ;)

Patrick Bourg - 5 years, 4 months ago

Thank you for posting a hard solution!!!

Elisa Abarquez - 3 years, 3 months ago
Rajat De
Jan 17, 2015

include <iostream>

using namespace std;

int main() { // your code goes here float a[10001]; a[0]=1; a[1]=2; a[2]=3; for(int i=3;i<10001;i++){ a[i] =a[i-1]+a[i-2]+a[i-3]; a[i]/=3; } cout<<a[10000]; }

Rajat, your algorithm does find the convergence, but unfortunately you will need to convert the resulting float (2.333333325 or something like that) back to a reduced fraction (a/b) in order to give the correct answer (a=7,b=3,a+b=10). See http://stackoverflow.com/questions/95727/how-to-convert-floats-to-human-readable-fractions for methods to do this. It is a non-trivial coding.

Richard Levine - 6 years, 4 months ago

0 pending reports

×

Problem Loading...

Note Loading...

Set Loading...