Let the sequence of real numbers { a n } n = 1 k be defined as a 1 = 3 , a 2 = 1 , and a n + 2 − a n a n + 1 a n + 2 = a n + a n + 1 for positive integers 1 ≤ n ≤ k − 2 . Find the largest possible positive integer value of k .
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.
Goddammit I got all these but I thought the question was asking for the greatest possible a_n
This one is tricky. Rearranging the recurrence, I found a computable form: a n + 2 = 1 − a n a n + 1 a n + a n + 1 so clearly we're looking for k such that a k − 1 a k − 2 = 1 . I wasn't sure how to proceed with this problem (since there doesn't seem to be a nice way to get the closed form solution of this sequence, so I just computed out a couple of sequences by hand. After a while, I started searching for common functions that produces 3 , and a quick google search took me to the trigonometric section. I tried the inverse sin function to no avail, but when I tried the inverse tan function: Bingo, all of the terms were some rational multiple of π !
This got me excited, and a quick search on trigonometric identities gave the following: tan ( b n + b n + 1 ) = 1 − tan ( b n ) tan ( b n + 1 ) tan ( b n ) + tan ( b n + 1 ) which has the same structure as our recurrence! This means that the sequence a n can be thought of as the injection of some sequence b n via the tan function! That is, let a n = tan ( b n ) , then a n + 2 = tan ( b n + 2 ) = 1 − tan ( b n ) tan ( b n + 1 ) tan ( b n ) + tan ( b n + 1 ) = tan ( b n + b n + 1 ) then, it's easy to see that letting b 1 = arctan ( 3 ) = 3 π and b 2 = arctan ( 1 ) = 4 π , the sequence b n + 2 = b n + 1 + b n can be transformed into a n via a n = tan ( b n ) .
But then b n is just the fibonacci sequence, so we can expression b n = f n − 2 b 1 + f n − 1 b 2 (it would be n-1 and n, but we started our index at one instead of zero). Now, this sequence terminates immediately 2 π ∣ b n , which occurs at k + 1 = 1 0 because b 1 0 = 2 3 1 π , so k = 9 .
Rearranging the given equation gives a n + 2 = 1 − a n a n + 1 a n + a n + 1 . This is exactly the recurrence satisfied by tan ( x ) . We know a 1 = tan ( 6 0 ) , a 2 = tan ( 4 5 ) , so a 3 a 4 a 5 a 6 a 7 a 8 a 9 = = = = = = = tan ( 6 0 + 4 5 ) = tan ( 1 0 5 ) tan ( 4 5 + 1 0 5 ) = tan ( 1 5 0 ) tan ( 1 0 5 + 1 5 0 ) = tan ( 2 5 5 ) = tan ( 7 5 ) tan ( 1 5 0 + 7 5 ) = tan ( 2 2 5 ) = tan ( 4 5 ) tan ( 7 5 + 4 5 ) = tan ( 1 2 0 ) tan ( 4 5 + 1 2 0 ) = tan ( 1 6 5 ) tan ( 1 2 0 + 1 6 5 ) = tan ( 1 0 5 ) However, if a n is defined for n = 1 0 , we would have a 1 0 = tan ( 1 6 5 + 1 0 5 ) = tan ( 2 7 0 ) which is undefined. Therefore, a n can only defined up until k = 9 .
it's a bit confusing that the question states " 1 ≤ n ≤ k − 2 " . I got n = 9 but I entered 11 as the question is asking about largest possible k
Problem Loading...
Note Loading...
Set Loading...
By rearranging the expression, we get
a n + 2 = 1 − a n a n + 1 a n + a n + 1
This looks like the formula
tan (A + B) = 1- tan A tanB tan A + tan B
Since a 1 = 3 , which is tan 6 0 ∘
and a 2 = 1 , which is tan 4 5 ∘
Therefore, a 3 = tan ( 6 0 ∘ + 4 5 ∘ ) = tan 1 0 5 ∘
Similarly, a 4 = tan ( 4 5 ∘ + 1 0 5 ∘ ) = tan 1 5 0 ∘
a 5 = tan ( 1 0 5 ∘ + 1 5 0 ∘ ) = tan 2 5 5 ∘
a 6 = tan ( 1 5 0 ∘ + 2 5 5 ∘ ) = tan 4 0 5 ∘
a 7 = tan ( 2 5 5 ∘ + 4 0 5 ∘ ) = tan 6 6 0 ∘
a 8 = tan ( 4 0 5 ∘ + 6 6 0 ∘ ) = tan 1 0 6 5 ∘
a 9 = tan ( 6 6 0 ∘ + 1 0 6 5 ∘ ) = tan 1 7 2 5 ∘
a 1 0 = tan ( 1 0 6 5 ∘ + 1 7 2 5 ∘ ) = tan 2 7 9 0 ∘
However, tan of odd multiples of 9 0 ∘ is not defined, and 2 7 9 0 ∘ is an odd multiple of 9 0 ∘ , so { a n } n = 1 k is defined only for 1 ≤ n ≤ 9
Therefore, k = 9