Difference between revisions of "1993 IMO Problems/Problem 1"
(Created page with "Let <math>f\left(x\right)=x^n+5x^{n-1}+3</math>, where <math>n>1</math> is an integer. Prove that <math>f\left(x\right)</math> cannot be expressed as the product of two non-co...") |
(→Solution) |
||
Line 9: | Line 9: | ||
From this fact, we have that there must exist a rational root of <math>f\left(x\right)</math>. The only candidates are <math>1</math>, <math>-1</math>, <math>3</math>, and <math>-3</math>, though. <math>f\left(1\right)=9\neq0</math>, <math>f\left(-1\right)=\pm4+3\neq0</math>, <math>f\left(3\right)=3^n+5\cdot3^{n-1}+3>0</math>, and <math>f\left(-3\right)=\pm\left(2\cdot3^n\right)+3\neq0</math>, so none of these work. Thus, there are no linear factors of <math>f\left(x\right)</math>. | From this fact, we have that there must exist a rational root of <math>f\left(x\right)</math>. The only candidates are <math>1</math>, <math>-1</math>, <math>3</math>, and <math>-3</math>, though. <math>f\left(1\right)=9\neq0</math>, <math>f\left(-1\right)=\pm4+3\neq0</math>, <math>f\left(3\right)=3^n+5\cdot3^{n-1}+3>0</math>, and <math>f\left(-3\right)=\pm\left(2\cdot3^n\right)+3\neq0</math>, so none of these work. Thus, there are no linear factors of <math>f\left(x\right)</math>. | ||
− | In other words, <math>f\left(x\right)</math> cannot be expressed as <math>g\left(x\right)h\left(x\right)</math> for polynomials <math>g\left(x\right)</math> and <math>h\left(x\right)</math> in <math>\mathbb{R}</math>. This means that | + | In other words, <math>f\left(x\right)</math> cannot be expressed as <math>g\left(x\right)h\left(x\right)</math> for polynomials <math>g\left(x\right)</math> and <math>h\left(x\right)</math> in <math>\mathbb{R}</math>. This means that <math>f\left(x\right)</math> cannot be expressed as the product of two non-constant polynomials with integer coefficients. |
Q.E.D. | Q.E.D. |
Revision as of 15:39, 28 July 2015
Let , where is an integer. Prove that cannot be expressed as the product of two non-constant polynomials with integer coefficients.
Solution
For the sake of contradiction, assume that for polynomials and in . Furthermore, let with if and with if . This gives that .
We have that , or . WLOG, let (and thus ). Since and divides but not , we need that . We can keep on going up the chain until we get that . Then, by equating coefficients once more, we get that . Taking the equation gives that . This implies that . Thus, the degree of is at least . However, because is a non-constant factor of , we have that the degree of is at most . Thus, the degree of \textit{is} , implying that the degree of is .
From this fact, we have that there must exist a rational root of . The only candidates are , , , and , though. , , , and , so none of these work. Thus, there are no linear factors of .
In other words, cannot be expressed as for polynomials and in . This means that cannot be expressed as the product of two non-constant polynomials with integer coefficients.
Q.E.D.