0
$\begingroup$

I was reading these solutions online http://campus.lakeforest.edu/trevino/Spring2019/Math331/Homework7Solutions.pdf for practice for an upcoming exam I have . But in question $2$ part $(a)$ the author uses a trick I'm not familiar with , he takes the derivative of the polynomial $x^5-12x^2+2$ , notes that the roots of said derivative are $0$ and $\sqrt[3]{\tfrac{24}{5}}$ and then from this deduces that there are at most 3 real roots to our original polynomial .

What theorem is the author invoking here ?

$\endgroup$
  • 4
    $\begingroup$ Looks like Rolle's Theorem. $\endgroup$ – Torsten Schoeneberg Jul 14 at 6:12
  • 2
    $\begingroup$ a lot of it is calculus: Rolle's theorem with MVT and IVT with FTA, conjugate pair theorem $\endgroup$ – user29418 Jul 14 at 6:14
  • $\begingroup$ @user29418 I know MVT stands for mean value theorem , what does IVT and FTA stand for though $\endgroup$ – excalibirr Jul 14 at 6:17
  • $\begingroup$ @user29418 I think IVT stands for intermediate value theorem ,not sure about FTA still though $\endgroup$ – excalibirr Jul 14 at 6:21
  • 2
    $\begingroup$ FTA is almost surely the Fundamental Theorem of Algebra $\endgroup$ – Rylee Lyman Jul 14 at 6:35
1
$\begingroup$

Just to expand on my comment, and to see that no other theorem (certainly not IVT or the Fundamental Theorem) need to be used here:

The standard version of Rolle's Theorem asserts that if a real-valued function is continuous on the interval $[a,b]$ and differentiable inside that interval, and $f(a) = f(b)$, then there is at least one $c \in (a,b)$ with $f'(c)=0$.

Now, since any polynomial $p$ is differentiable and hence continuous everywhere, Rolle's Theorem asserts that between any two of its zeroes (taking the role of $a$ and $b$ above), its derivative $p'$ must also have a zero. This easily generalises to:

If $p$ has $n$ different zeroes, then $p'$ has at least $n-1$ different zeroes

(namely, at least one zero of $p'$ between the first and second zero of $p$, then at least one between the second and third, ..., and at least one between the $n-1$th and the $n$th zero of $p$, where by "first", "second" etc. I just order the zeroes from left to right on the real number line).

Which, when you think about it for a moment, implies that:

If $p'$ has $n-1$ different zeroes, then $p$ cannot have more than $n$ different zeroes

which is what is used in your case, where $n=3$.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.