22
$\begingroup$

This question already has an answer here:

I can't solve this problem:

Suppose $f$ is polynomial function of even degree $n$ with always $f\geq0$.

Prove that $f+f'+f''+\cdots+f^{(n)}\geq 0$.

$\endgroup$

marked as duplicate by Martin, Amzoti, Lord_Farin, Start wearing purple, Chris Godsil Jun 11 '13 at 21:39

This question has been asked before and already has an answer. If those answers do not fully address your question, please ask a new question.

44
$\begingroup$

Let $g=f+f'+\cdots+f^{(n)}$ and let $h(x)=e^{-x}g(x)$. Note that $f^{(n+1)}=0$, so $$h'(x)=e^{-x}(g'(x)-g(x))=-e^{-x}f(x)\le 0,$$ i.e. $h$ is decrearing on $\mathbb R$. Since $g$ is a polynomial, $\lim\limits_{x\to+\infty}h(x)=0$. It follows that $h\ge 0$, and hence $g\ge 0$.

$\endgroup$
  • $\begingroup$ Why lim h(x) goes to zero? $\endgroup$ – user75086 Jun 8 '13 at 11:24
  • 2
    $\begingroup$ @user75086: As $x\to +\infty$, the exponential function $e^x$ tends to $\infty$ much faster than any polynomial function. $\endgroup$ – 23rd Jun 8 '13 at 11:26
37
$\begingroup$

Alternatively, by a standard Maximum Value Theorem argument, the even-degree polynomial $g=f+f'+\dots+f^{(n)}$ has a global minimum at some point $a$. Then $g(a)=f(a)+g'(a)\ge 0$, so $g\ge g(a)\ge 0$.

$\endgroup$
  • 2
    $\begingroup$ It's nice and more natural than my argument. +1. $\endgroup$ – 23rd Jun 8 '13 at 11:19
  • $\begingroup$ why $g(a)=f(a)+g'(a)\ge 0$? $\endgroup$ – user75086 Jun 8 '13 at 11:55
  • 1
    $\begingroup$ @user75086: Because $g'(a)=0$ and $f(a)\ge 0$. $\endgroup$ – Ted Shifrin Jun 8 '13 at 12:16

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