Take the 2-minute tour ×
MathOverflow is a question and answer site for professional mathematicians. It's 100% free, no registration required.

Suppose you have two polynomials $P$ and $Q$ with integer coefficients. Let their GCD (more specifically, the smallest integer multiple of the GCD in $\mathbb Q$ such that all the coefficients are integral) be $D$.

If, for a particular value $x_0$, we have that $P(x_0)$ and $Q(x_0)$ are both integers, does it follow that $D(x_0)$ is an integer? Does $D(x_0)$ divide $P(x_0)$ and $Q(x_0)$?

share|improve this question
2  
@Felipe, whatever is meant by gcd, I don't think the OP is assuming $D$ is monic. Consider $P=Q=2x$. Then you'd better take $D=2x$ as well, otherwise $x_0=1/2$ would be a cheap counterexample. –  Barry Cipra Apr 18 '13 at 22:24
4  
However it's defined (monic or not), surely you must have $P/D$ and $Q/D$ polynomials with integer coefficients. What do you take for $2x$ and $2x^2+x$? Either it's a counterexample, or you aren't requiring the quotients to be in $\mathbb{Z}[x]$. –  Zack Wolske Apr 18 '13 at 22:43
1  
GCD should be considered in $\mathbb{Z}[x]$ as Zack pointed out. –  i707107 Apr 19 '13 at 0:40
2  
$\mathbb{Z}[x]$ is not PID, but it is a UFD. So, it will be clear if we think GCD as in UFD. –  i707107 Apr 19 '13 at 2:09
2  
Then Zach already gave a counterexample... –  Felipe Voloch Apr 19 '13 at 3:21
show 6 more comments

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Browse other questions tagged or ask your own question.