Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

Suppose I have $f(x)A+g(x)B+h(x)C \ge 0$. Here $A,B,C$ can be positive or negative and $f,g,h$ are nonnegative. I would like to obtain a condition for $f,g,$ and $h$ such that $f'(x)A+g'(x)B+h'(x)C \ge 0$. I will appreciate any substantial comments.

share|cite|improve this question
Unless you tell us more about $f$, $g$, $h$ and the $x$-domain $I$ where all this should hold the condition $f'(x)A+g'(x)B+h'(x)C \ge 0$ cannot be transformed into something simpler. If $I$ is a compact interval then the assumption $f(x)A+g(x)B+h(x)C \ge 0$ is useless, because by adding a suitable constant to $f$ you always can force it to hold without changing the derivative(s). – Christian Blatter Jul 9 '11 at 15:02
$I$ is a compact interval, and $A, B, C $are given. – user12847 Jul 9 '11 at 15:39

As the problem is currently stated, $A$, $B$, $C$, and $f$, $g$, $h$ are unfortunately not relevant. Let $$W(x)=Af(x)+Bg(x)+Ch(x).$$

The problem states that $W(x)\ge 0$, presumably for all $x$, and asks for conditions under which $W'(x) \ge 0$ for all $x$.

The condition $W(x) \ge 0$ cannot be of much help. We could ask for $W(x)$ to be non-decreasing, but that is really only a minor restatement of $W'(x)\ge 0$. Apart from that sort of thing, there is no nice condition on a general function that will ensure a non-negative derivative. And despite the apparent complexity of $W(x)$, there are no conditions on it apart from $W(x) \ge 0$.

share|cite|improve this answer
Thank you. I mean, given $A, B$, and $C$, what are the restrictions for $f,g,$ and $h$ that are sufficient for the second inequallity to hold? – user12847 Jul 9 '11 at 15:41
@user12847: At the level of generality you seek, even if we put $A=B=C=1$ we can say nothing useful. Even if we put $A=1$, $B=C=0$ we can say nothing useful. I am sure this comes from a real problem, with considerable information about the shapes of the various functions. With (a lot of) further information, there is likely to be an answer. – André Nicolas Jul 9 '11 at 15:53
Let us say $C=0$. Then $fA+gB \ge 0.$ Suppose $A, B \ge 0.$ Then it will give $B \ge (-f/g) A$. Then $f'A -g'(f/g) A \ge 0$ is a sufficient condition. This level of condition is fine. – user12847 Jul 9 '11 at 16:05
An additional hint to make you see that @user6312's answer is indeed the crux of the matter: the condition $W(x)\ge0$ is invariant if one multiplies $f(x)$, $g(x)$ and $h(x)$ by $\mathrm{e}^{-ux}$, for every fixed $u$, right? But the condition $W'(x)\ge0$ becomes $W'(x)-uW(x)\ge0$. This becomes a huge problem if $u\gg1$. – Did Jul 9 '11 at 16:06
I am not asking a general condition. I am asking conditions given A, B, and C fixed. – user12847 Jul 9 '11 at 16:14

Your Answer


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

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