Conside The Fibonacci sequence is defined by f 1 = 1 , f 2 = 1 , f n = f n − 1 + f n − 2 for n ≥ 3 . We have that f 1 3 = 2 3 3 . Consider a sequence such that
g 1 = 1 , g 2 = x , g n = g n − 1 + g n − 2 for n ≥ 3 .
Determine the sum of all positive integer values of x ≥ 2 for which 2 3 3 is a term of the sequence g .
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.
Is there any way to get this answer without brute forcing it, or at least an explanation as to why the ones that work work? Is it truly just a coincidence?
Log in to reply
If anyone is still here, how could this problem be approached for numbers larger than 233? Does finding a generalised closed form help?
Lemma : It is clear that for positive integer n ≥ 4 , g n = f n − 3 + f n − 2 x . Proof : We use strong induction on n .
Base case - We need to prove this for n = 4 and n = 5 .
If n = 4 , g 4 = g 2 + g 3 = g 2 + ( g 1 + g 2 ) = 1 + 2 x . Since f 1 = 1 and f 2 = 2 , we have g 4 = f 1 + f 2 x as required.
If n = 5 , g 5 = g 3 + g 4 = ( g 1 + g 2 ) + ( 1 + 2 x ) = 2 + 3 x . Since f 2 = 2 , f 3 = 3 we have g 5 = f 2 + f 3 x as required.
Induction step - For arbitrary k ≥ 5 , we take g k = f k − 3 + f k − 2 x and g k − 1 = f k − 4 + f k − 3 x . We need to show g k + 1 = f k − 2 + f k − 1 x .
By definition of g we have g k + 1 = g k + g k − 1 g k + 1 = f k − 3 + f k − 2 x + f k − 4 + f k − 3 x Refactoring, g k + 1 = ( f k − 3 + f k − 4 ) + ( f k − 2 + f k − 3 ) x By the definition of Fibonacci numbers, we have g k + 1 = f k − 2 + f k − 1 x as required.
So if the sequence contains 2 3 3 then for some n , we will have 2 3 3 = f n − 3 + f n − 2 x . Then, for n ≥ 4 by the lemma we have x = f n − 2 2 3 3 − f n − 3 The first 1 2 terms of the Fibonacci sequence are ( 1 , 2 , 3 , 5 , 8 , 1 3 , 2 1 , 3 4 , 5 5 , 8 9 , 1 4 4 , 2 3 3 ) - clearly since x is a positive integer we need not be concerned with the cases where f n − 3 ≥ 2 3 3 . Checking each pair of consecutive terms gives us the values x = 2 , 4 6 , 7 7 , 1 1 6 But our equation only applied for n ≥ 4 - we observe that it is also possible that g 2 or g 3 could be 2 3 3 . If g 2 = 2 3 3 then x = 2 3 3 ; if g 3 = 2 3 3 then we have 1 + x = 2 3 3 and x = 2 3 2 . So we must add to our earlier solutions the values x = 2 3 2 , 2 3 3 .
Hence our answer is 2 + 4 6 + 7 7 + 1 1 6 + 2 3 2 + 2 3 3 = 7 0 6 .
Great Lemma. This is useful in helping us understand linear recurrences which satisfy g n + 2 = g n + 1 + g n .
A simpler / direct proof uses the characterization of solutions, which must be of the form g n = A α n + B β n , where A , B are constants and α , β are roots of x 2 = x + 1 .
You have a lovely looking solution here. Even though it is same as "brute" method used by Jimmy K, it looks better.
CM do you have a solution along those lines? I tried that at first but I didn't see how to get 233 out of the resulting expression
Log in to reply
Solving the recurrence relation g n = g n − 1 + g n − 2 , we look for solutions g n = λ n , leading to the equation λ 2 − λ − 1 = 0 . Thus g n = A α n − 1 + B β n − 1 for n ≥ 1 , where α = 2 1 ( 1 + 5 ) and β = 2 1 ( 1 − 5 ) . The conditions g 1 = 1 and g 2 = x give A B = = 2 1 + 2 5 1 ( 2 x − 1 ) = 5 1 ( x − β ) 2 1 − 2 5 1 ( 2 x − 1 ) = 5 1 ( − x + α ) so g n = 5 1 [ x ( α n − 1 − β n − 1 ) + ( α n − 2 − β n − 2 ) ] n ≥ 1 using the fact that α β = − 1 . But, putting x = 2 we see that f n = C α n − 1 + D β n − 1 where C D = = 5 1 ( 2 − β ) = 5 1 ( 1 + α ) = 5 1 α 2 5 1 ( − 2 + α ) = 5 1 ( − 1 − β ) = − 5 1 β 2 so that f n = 5 1 ( α n + 1 − β n + 1 ) and hence g n = x f n − 2 + f n − 3 as required.
This question is a little strange, since the Fibonacci numbers are usually defined as starting 0 , 1 , 1 , 2 , 3 , 5 , … . If we start with f − 1 = 0 and f 0 = 1 , then the above formula for f n is valid for n ≥ − 1 , and the above formula relating g n to f n − 2 and f n − 3 is valid for n ≥ 2 , and so Jeremy K. needs no special argument to deal with g 2 and g 3 .
I do get the induction part, but why was it necessary when you will ultimately have to brute your way through the values?
The values of x can be obtained by listing the terms in the sequence having first term=1 and second term=x.
These are:
1,x,(x+1),(2x+1),(3x+2),(5x+3),(8x+5),(13x+8),(21x+13),(34x+21),(55x+34),(89x+55).
Since, x>=2, we stop at 89x+55.
We then equate the terms (except of course the first term which is 1) to 233 in order to find positive integer values of x.
We obtain the values:
233, 232, 116, 77, 46, 2.
Therefore, the sum of all the possible values of x is 233+232+116+77+46+2=706. :)
Our sequence will be: 1 , x , 1 + x , 1 + 2 x , 2 + 3 x , 3 + 5 x , 5 + 8 x , 8 + 1 3 x
, 1 3 + 2 1 x , 2 1 + 3 4 x , 3 4 + 5 5 x , 5 5 + 8 9 x .
By trial, we find 6 possible values of x that satisfy the given condition: 233,232,116,77,46,2.
The answer is 233+232+116+77+46+2=706
The sequence
g
is defined as follows:
1
,
x
,
1
+
x
,
1
+
2
x
,
2
+
3
x
,
3
+
5
x
,
5
+
8
x
,
8
+
1
3
x
,
1
3
+
2
1
x
,
2
1
+
3
4
x
,
3
4
+
5
5
x
,
5
5
+
8
9
x
We can stop here because any term above this will necessarily be greater than
2
3
3
for
x
≥
2
.
For each of these terms of g , we desire to find a positive integer solution of x for g n = 2 3 3 . We could brute force it quite easily, giving us values of 2 3 3 , 2 3 2 , 1 1 6 , 7 7 , 4 6 , 2 whose sum is 7 0 6 .
Computing the first few terms of g in terms of x , we have
g 1 = 1
g 2 = x
g 3 = 1 + x
g 4 = 1 + 2 x
g 5 = 2 + 3 x
g 6 = 3 + 5 x
g 7 = 5 + 8 x
g 8 = 8 + 1 3 x
g 9 = 1 3 + 2 1 x
g 1 0 = 2 1 + 3 4 x
g 1 1 = 3 4 + 5 5 x
g 1 2 = 5 5 + 8 9 x .
Note that when x = 2 , g 1 2 = 2 3 3 . Thus, x = 2 is a solution. Since we must have x ≥ 2 , when a > 1 2 , then g a cannot equal 2 3 3 . (If x ≥ 2 ). (Note that as x increases, g 1 2 increases)
If 233 is a term of the sequence g , then g m must equal 233 for some m . Thus, we can solve the equations g m = 0 for each m ≤ 1 2 . g 2 = 2 3 3 ⇒ x = 2 3 3 g 3 = 2 3 3 ⇒ 1 + x = 2 3 3 ⇒ x = 2 3 2 g 4 = 2 3 3 ⇒ 1 + 2 x = 2 3 3 ⇒ x = 1 1 6 g 5 = 2 3 3 ⇒ 2 + 3 x = 2 3 3 ⇒ x = 7 7 g 6 = 2 3 3 ⇒ 3 + 5 x = 2 3 3 ⇒ x = 4 6 g 7 = 2 3 3 ⇒ 5 + 8 x = 2 3 3 ⇒ x = 2 8 . 5 ∈ Z g 8 = 2 3 3 ⇒ 8 + 1 3 x = 2 3 3 ⇒ x = 1 3 2 2 5 ∈ Z g 9 = 2 3 3 ⇒ 1 3 + 2 1 x = 2 3 3 ⇒ x = 2 1 2 2 0 ∈ Z g 1 0 = 2 3 3 ⇒ 2 1 + 3 4 x = 2 3 3 ⇒ x = 1 7 1 0 6 ∈ Z g 1 1 = 2 3 3 ⇒ 3 4 + 5 5 x = 2 3 3 ⇒ x = 5 5 1 9 9 ∈ Z g 1 2 = 5 5 + 8 9 x ⇒ x = 2 .
Note that a ∈ Z means that a is not an integer. Since we are looking for integer solutions, these values are to be excluded.
Thus, our solutions are 2 3 3 , 2 3 2 , 1 1 6 , 7 7 , 4 6 , 2 , and their sum is 7 0 6
Writing out the first 1 2 terms of sequence g ; 1 , x , x + 1 , 2 x + 1 , 3 x + 2 , 5 x + 3 , 8 x + 5 , 1 3 x + 8 , 2 1 x + 1 3 , 3 4 x + 2 1 , 5 5 x + 3 4 , 8 9 x + 5 5 Because x ≥ 2 , we see that the only way that 2 3 3 will be a term of sequence g is when either the 2 nd term or every term after it until the 1 2 th term is equal to 2 3 3 . When 2 3 3 is equal to the 1 2 th term, we get x = 2 . Using the similar method for the 2 nd though 1 1 th terms, we get x = 2 , 4 6 , 7 7 , 1 1 6 , 2 3 2 , 2 3 3 .
2 + 4 6 + 7 7 + 1 1 6 + 2 3 2 + 2 3 3 = 7 0 6
Problem Loading...
Note Loading...
Set Loading...
Using the recursion, g 3 = x + 1 , g 4 = 2 x + 1 , g 5 = 3 x + 2 , g 6 = 5 x + 3 , g 7 = 8 x + 5 , g 8 = 1 3 x + 8 , g 9 = 2 1 x + 1 3 , g 1 0 = 3 4 x + 2 1 , g 1 1 = 5 5 x + 3 4 , and g 1 2 = 8 9 x + 5 5 .
Since g 2 > g 1 > 0 and g n = g n − 1 + g n − 2 for n ≥ 3 we can show that g n is a strictly increasing sequence. Then, since x ≥ 2 , we have that g n > g 1 2 = 8 9 x + 5 5 ≥ 2 3 3 for all n ≥ 1 3 . So, we only need to find integers x ≥ 2 such that g n = 2 3 3 for some positive integer n ≤ 1 2 .
We have just 1 2 easy cases to check:
g 1 = 1 = 2 3 3
g 2 = x = 2 3 3 ⇝ x = 2 3 3
g 3 = x + 1 = 2 3 3 ⇝ x = 2 3 2
g 4 = 2 x + 1 = 2 3 3 ⇝ x = 1 1 6
g 5 = 3 x + 2 = 2 3 3 ⇝ x = 7 7
g 6 = 5 x + 3 = 2 3 3 ⇝ x = 4 6
g 7 = 8 x + 5 = 2 3 3 ⇝ x = 2 5 7
g 8 = 1 3 x + 8 = 2 3 3 ⇝ x = 1 3 2 2 5
g 9 = 2 1 x + 1 3 = 2 3 3 ⇝ x = 2 1 2 2 0
g 1 0 = 3 4 x + 2 1 = 2 3 3 ⇝ x = 1 7 1 0 6
g 1 1 = 5 5 x + 3 4 = 2 3 3 ⇝ x = 5 5 1 9 9
g 1 2 = 8 9 x + 5 5 = 2 3 3 ⇝ x = 2
Hence, the sum of all positive integers x ≥ 2 such that g n = 2 3 3 for some integer n is 2 3 3 + 2 3 2 + 1 1 6 + 7 7 + 4 6 + 2 = 7 0 6 .