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

Problem: Let $f,g: [0, + \infty[ \to \mathbb{R}$ be two convex functions of Class $C^2$. Assume that $$ f(0) \geq 0, g(0) \geq 0 \text{ and } f'(0) \geq 0, g'(0) \geq 0 \tag{!}$$ Show that $fg$ is convex

My approach: I don't really understand what to do with the given information tagged as (!). I do however know the following:

I focus on $f$, the same will apply for $g$. Since $f$ is a convex function of Class $C^2$ and we have $f'(0) \geq 0$ there must exists some positive $\delta_1>0$ for which we have $$f': [0, \delta_1] \to \mathbb{R} \text{ is increasing } \implies f'' \geq 0, \ \forall x \in [0, \delta_1] $$

Analogous for $g$ with some different $\delta_2 >0$. Define $\delta =\min(\delta_1,\delta_2)$ such that both $f$ and $g$ are on the Intervall $I=[0,\delta]$ simultaneously convex meaning: $$f'' \geq 0 \text{ and } g''>0,  \forall x \in [0,\delta] $$

The above was merely some pre work I did to understand the problem better, not sure if it was necessarily.

For the 2nd part I define $h: I=[0, \delta] \to \mathbb{R}, \ h \mapsto fg \in C^2$ because it is a composition of $C^2$ functions. Now I differentiate $h$ twice: $$h'=f'g+g'f \implies h''=f''g+g'f'+g''f+f'g' \\ \implies h''=f''g+2f'g'+g''f \geq 0, \forall x \in I $$

Which would show that $h$ is convex. It looks correct to me but I'd like to hear some opinions if my approach was correct or if I have some flawed thinking at some points. Please comment/answer to highlight my mistakes or suggest improvisations.

share|cite|improve this question
up vote 2 down vote accepted

You have $h'' = f''g + 2f'g' + g''f$, but you haven't explained why $f''g + 2f'g' + g''f \geq 0$. You can assert this if you know that $f''$, $g''$, $f'$, $g'$, $f$, and $g$ are all positive on $[0,\infty)$. You know that $f''$ and $g''$ are positive by convexity. The information in (!) is supposed to help you show that $f'$, $g'$, $f$, and $g$ are all positive. Then you will be done.

share|cite|improve this answer
Thanks for your answer @AlexZorn, I was trying to empathize that the 2nd derivatives of $f,g$ are strictly positive on the interval I have defined. But that seems to have been a redundant step because, as you said, they are convex. I do however believe that my defined pre work shows that $f',g',f,g$ are positive on the interval $I=[0, \delta]$ – Spaced Apr 30 '14 at 14:43

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.