Let p , q and r be the roots of the equation x 3 + 3 x = 2 x 2 + 4 , and denote S n = p n + q n + r n . Find the sum of all positive integers m satisfying ∣ S m ∣ = 2 .
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.
How do you know that there's no solution for m > 1 2 ?
Log in to reply
I recognized that I had major flaws in my argument justifying that the solutions 1 , 2 , 3 , 6 , 7 , 1 1 are the only ones. I have modified my answer to take care of those caveats.
Great way to do the problem, especially to get the induction between the nth term with (n-4)th and (n-5)th term. I believe there are two small errors, one is the induction should be S(n) = 5S(n-4)+4S(n-5), another minor one is S(4) = 18, not -10.
Log in to reply
Thanks a lot for pointing out the mistakes. I also recognized that there was a major flaw in my previous argument regarding why the numbers 1 , 2 , 3 , 6 , 7 , 1 1 are going to be the only solutions. I have modified my answer to take care of those points.
Log in to reply
You are welcome. Yeah, I was thinking about that argument of no more solutions after m=11, your current way is a great way to go. Like you said, we just need a block of 5 S (from 7 to 11) that is all positive, then can apply induction to show all future S are greater than 2. The recursion you get is crucial to carry out the induction, as the original recursion has negative coefficients, which is not good.
The cubic f ( x ) = x 3 − 2 x 2 + 3 x − 4 has one real root p and two complex roots q , r = q ⋆ . Since f ( 1 . 6 5 ) < 0 < f ( 1 . 6 6 ) , we deduce that 1 . 6 5 < p < 1 . 6 6 . Also 4 = p q r = p ∣ q ∣ 2 , and so ∣ q ∣ = p 2 , so that 1 . 6 6 2 < ∣ q ∣ < 1 . 6 5 2 . If we write q = ∣ q ∣ e i θ then S n = p n + q n + r n = p n + 2 ∣ q ∣ n cos n θ = ∣ q ∣ n [ ( ∣ q ∣ p ) n + 2 cos n θ ] Since ( ∣ q ∣ p ) n > ( 1 . 6 5 2 1 . 6 5 ) n = ( 2 1 . 6 5 1 . 5 ) n and ( 2 1 . 6 5 1 . 5 ) 1 2 > 2 , it is clear that S n > 0 for all n ≥ 1 2 , and moreover that S n ≥ ∣ q ∣ n [ ( ∣ q ∣ p ) n − 2 ] ≥ ( 1 . 6 6 2 ) n [ ( 2 1 . 6 5 1 . 5 ) n − 2 ] n ≥ 1 2 . Hence we deduce that S n ≥ ( 1 . 6 6 2 ) n [ ( 2 1 . 6 5 1 . 5 ) n − 2 ] ≥ ( 1 . 6 6 2 ) 1 3 [ ( 2 1 . 6 5 1 . 5 ) 1 3 − 2 ] > 3 8 for all n ≥ 1 3 . Certainly, then, ∣ S n ∣ > 2 for all n ≥ 1 3 . It is now just a matter of checking that the recurrence relation S n + 3 − 2 S n + 2 + 3 S n + 1 − 4 S n = 0 S − 1 = 4 3 , S 0 = 3 , S 1 = 2 gives S 1 = S 3 = S 7 = S 1 1 = 2 and S 2 = S 6 = − 2 , and that no other values of S n for 1 ≤ n ≤ 1 2 are equal to 2 or − 2 .
Thus the answer is 1 + 2 + 3 + 6 + 7 + 1 1 = 3 0
You always come up with "out of this world" approaches.
Problem Loading...
Note Loading...
Set Loading...
The observation that I found to be the key to solving the problem was that after some algebraic manipulations, the equation can be reduced to x 5 − 5 x − 4 = 0 This leads to the following recursive relation, S n = 5 S n − 4 + 4 S n − 5 , n ≥ 6 Now, one can check it manually that S 1 = 2 , S 2 = − 2 , S 3 = 2 , S 4 = 1 8 . This implies that every S n is an integer. Note that we do not need to calculate any more S n 's manually, and we can use the recursive relation to our advantage to tell which n 's are going to give ∣ S n ∣ = 2 . Now, note that apart from 1 , 2 , 3 , the only other n 's for which ∣ S n ∣ = 2 , are going to be 6 (from { 1 , 2 } ), 7 (from { 2 , 3 } ), and 1 1 (from { 6 , 7 } ). This can be seen by an use of the recursive relation, as the relation implies that if both of S n − 4 , S n − 5 are positive, S n is going to be higher than 2 . Since every S n , for n ≥ 1 2 can be determined, using the recursion, from the S n 's with n ≥ 7 , and since every S 7 > 0 , S n > 2 ∀ n ≥ 1 2 . Thus the answer is 3 0 .