How do you find the roots for f(x) = 15x^12 + 41x^9 + 13x^2 –10 using the fundamental theorem of algebra?

1 Answer
Jun 5, 2016

The FTOA tells us it has 12 zeros but does not help us find them...

Explanation:

The fundamental theorem of algebra tells you that a polynomial of degree n > 0 in one variable has a Complex (possibly Real) zero.

A corollary of this, often stated as part of the FTOA is that a polynomial of degree n > 0 in one variable will have exactly n Complex (possibly Real) zeros counting multiplicity.

To see why this follows, note that if we have one zero x_1, then f(x) will be divisible by (x-x_1) resulting in a polynomial of degree n - 1. If n - 1 > 0 then this will have a zero x_2, etc.

In our example:

f(x) = 15x^12+41x^9+13x^2-10

is of degree 12, so has exactly 12 zeros counting multiplicity.

That's all the FTOA tells you. It does not help you find the zeros.

color(white)()
What else can we find out about the zeros of this polynomial?

Note that the signs of the coefficients follow the pattern:

+ + + -

With one change of sign, that means that there will be exactly one positive Real zero.

Reversing the sign on the term of odd degree we get the pattern:

+ - + -

With three changes of sign, that means that there will be either 1 or 3 negative Real zeros.

We can try to find rational zeros using the rational root theorem:

Any zero of f(x) must be expressible in the form p/q for integers p, q with p a divisor of the constant term -10 and q a divisor of the coefficient 15 of the leading term.

That means that the only possible rational zeros are:

+-1/15, +-2/15, +-1/5, +-1/3, +-2/5, +-2/3 +-1, +-5/3, +-2, +-5, +-10

None of these works, so f(x) has no rational zeros.

Ultimately, we are stuck with using numerical methods (e.g. Durand-Kerner) to find approximations for the zeros:

x~~-1.38923

x~~0.735661

x~~-0.775742+-0.203247i

x~~-0.487897+-0.736584i

x~~0.14955+-0.928279i

x~~0.714459+-1.21625i

x~~0.726412+-0.459408i