Suppose f and g are polynomials with integer coefficients such that f ( 0 ) = 0 , f has degree at least two, and f ( g ( x ) ) = f ( x ) + ( x 6 + 3 x 5 − 6 x 3 + 6 x 2 ) . Find the last three digits of the sum of absolute values of all distinct possible values of f ( 1 0 ) .
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.
Very nice solution! The case-by-case analysis may seem long, but the cases are actually very easy.
How do you get from x − y dividing x n − y n , to g − x dividing f ( g ) − f ( x ) ?
Log in to reply
If f ( x ) = a n x n + a n − 1 x n − 1 + ⋯ + a 1 x + a 0 , then f ( x ) − f ( y ) = = = ∑ m = 0 n a m x m − ∑ m = 0 n a m y m ∑ m = 1 n a m ( x m − y m ) ( x − y ) ∑ m = 1 n a m ( x m − 1 + x m − 2 y + ⋯ + x y m − 2 + y m − 1 ) and so x − y divides f ( x ) − f ( y ) . Now put y = g ( x ) .
Note that g ( x ) − x ∣ f ( g ( x ) ) − f ( x ) = x 6 + 3 x 5 − 6 x 3 + 6 x 2 . Further, note that x 4 + 3 x 3 − 6 x + 6 is irreducible, limiting the possible values of g ( x ) − x .
Case 1: g has degree at least 2:
Note that the degree of f ( g ( x ) ) − f ( x ) is equal to the degree of f ( g ( x ) ) , since that is strictly larger than the degree of f ( x ) .
Since f has degree at least 2, g has degree at most 3. However, there is no degree 3 factor of x 6 + 3 x 5 − 6 x 3 + 6 x 2 , so f has degree 3 and g has degree 2. We know that g is either x + x 2 or x − x 2 , since there are only two factors of x 6 + 3 x 5 − 6 x 3 + 6 x 2 of degree 2, namely x 2 and − x 2 .
Now, it is easy to substitute those two expressions into f ( x ) = A x 3 + B x 2 + C x to determine that only the first works, giving a unique solution f ( x ) = x 3 − 3 x 2 + 6 x , with f ( 1 0 ) = 7 6 0 .
Case 2: g is constant. g ( x ) = k ∀ x .
Then k is an integer such that k 6 + 3 k 5 − 6 k 3 + 6 k 2 = 0 , so k = 0 . This gives f ( x ) = − ( x 6 + 3 x 5 − 6 x 3 + 6 x 2 ) , with ∣ f ( 1 0 ) ∣ = 1 2 9 4 6 0 0 .
Case 3: g is linear. Say g ( x ) = p x + q .
Note that ( p − 1 ) x + q ∣ x 6 + 3 x 5 − 6 x 3 + 6 x 2 , so either p = 1 , or p − 1 = ± 1 . But p = 0 , so p = 2 in this case.
Case 3.1: p = 2 .
Then x + q ∣ x 6 + 3 x 5 − 6 x 3 + 6 x 2 , so q = 0 . It is easy to see, considering the coefficient of x 6 , that g ( x ) = 2 x has no solution.
Case 3.2: p = 1 .
Then the degree of f ( x + q ) − f ( x ) is one less than the degree of f ( x ) , so f is of degree 7. However, this gives a leading coefficient of f ( x + q ) − f ( x ) equal to seven times the leading coefficient of f , which is patently impossible.
Hence, the required sum is 1 2 9 4 6 0 0 + 7 6 0 = 1 2 9 5 3 6 0 , giving an answer of 3 6 0 .
This is a good solution, though if all the details were included it would have been quite lengthy.
Oh man. I found those two cases quickly, proving those were the only two cases, and got 760 + 400 = 1160. But it didn't accept 160, and then I spent about 10 more hours going over my algebra over and over trying to see where I might have missed a possible solution. :(
A hint would have been useful here (I was even thinking of posting a new thread asking for a hint as to whether there is a third solution I missed)
We can see that the degree of 2 sides must be equal. In the case deg(f) >= 6: we have f = (-x^6 +3x^5-6x^3+6x^2) In the case def(f) < 6 we have two cases of deg(f): 2 or 3 Solve we have f(x) = x^3 - 3x^2 + 6x
So that, result is 360
actually if f(x)=x^3 - 3x^2 + 6x then f(10)=760
Problem Loading...
Note Loading...
Set Loading...
Since x − y divides x n − y n for all positive integers n , we deduce that g ( x ) − x divides f ( g ( x ) ) − f ( x ) = = x 6 + 3 x 5 − 6 x 3 + 6 x 2 x 2 ( x 4 + 3 x 3 − 6 x + 6 ) and this quartic polynomial is irreducible over the integers (using Eisenstein's criterion with p = 3 ). If g ( x ) − x had degree m ≥ 4 , then g ( x ) would have the same degree, and hence f ( g ( x ) ) would have degree at least 2 m ≥ 8 . But this would force f ( x ) to have the same degree as f ( g ( x ) ) , which is impossible. Thus g ( x ) − x has degree at most 3 , so is coprime to x 4 + 3 x 3 − 6 x + 6 , and hence it divides x 2 . There are 6 options to consider:
If g ( x ) − x = 1 , then g ( x ) = x + 1 . However, if f ( x ) has leading term a x n , then f ( x + 1 ) − f ( x ) has leading term n a x n − 1 , and hence f ( x + 1 ) − f ( x ) cannot be equal to a monic polynomial of degree 6 . This case is not possible.
If g ( x ) − x = − 1 , then g ( x ) = x − 1 . But, if f ( x ) has leading term a x n , f ( x − 1 ) − f ( x ) has leading term − n a x n − 1 , and hence f ( x − 1 ) − f ( x ) cannot be a monic polynomial of degree 6 . This case is not possible.
If g ( x ) − x = x , then g ( x ) = 2 x . But. if f ( x ) has leading term a x n , f ( 2 x ) − f ( x ) has leading term ( 2 n − 1 ) a x n , and so cannot be equal to a monic polynomial of degree 6 . This case is not possible.
We can have g ( x ) − x = − x , so that g ( x ) = 0 and f ( x ) = − ( x 6 + 3 x 5 − 6 x 3 + 6 x 2 ) . This case gives f ( 1 0 ) = − 1 2 9 4 6 0 0 .
If g ( x ) − x = x 2 , we have g ( x ) = x 2 + x and therefore f ( x ) must be cubic. Chasing coefficients, it follows that f ( x ) = x 3 − 3 x 2 + 6 x . This gives f ( 1 0 ) = 7 6 0 .
If g ( x ) − x = − x 2 , we have g ( x ) = x − x 2 and therefore f ( x ) must be cubic. However, if f ( x ) has leading term a x 3 , then f ( x − x 2 ) − f ( x ) = − a x 6 + 3 a x 5 + ⋯ , and it is clear that no value of a exists so that f ( x − x 2 ) − f ( x ) = x 6 + 3 x 5 − 6 x 3 + 6 x 2 . This case is also impossible.
Thus there are only two valid cases, and since ∣ ∣ − 1 2 9 4 6 0 0 ∣ ∣ + ∣ ∣ 7 6 0 ∣ ∣ = 1 2 9 5 3 6 0 so the required answer is 3 6 0 .