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

How to show the following:

If $f:\mathbb R^d \rightarrow \mathbb R$ is convex

then its directional derivative is sublinear?

Thank you...

share|cite|improve this question
up vote 3 down vote accepted

I assume that you mean that the directional derivative is sublinear with respect to the direction? In that case we must prove that

  1. $f'(x;\alpha t) = \alpha f'(x; t)$
  2. $f'(x;t + s) \leq f'(x;t) + f'(x;s)$

Proof:

  1. $$f'(x;\alpha t) = \lim_{h \rightarrow 0 } \frac{f(x+h\alpha t) - f(x)}{h} = \alpha \lim_{h \rightarrow 0 } \frac{f(x+(h\alpha) t) - f(x)}{h\alpha} = \alpha f'(x;t)$$

  2. $$ f'(x;t + s) = \lim_{h \rightarrow 0 } \frac{f(x+h(t+s)) - f(x)}{h} \\ \\= \lim_{h \rightarrow 0 } \frac{f(\frac{1}{2}(x+2ht) + \frac{1}{2}(x+2hs)) - \frac{1}{2}f(x) - \frac{1}{2}f(x)}{h} \\ \leq \lim_{h \rightarrow 0 } \frac{f((x+2ht)) - f(x)}{2h} + \lim_{h \rightarrow 0 } \frac{f((x+2hs)) - f(x)}{2h} \\ = f'(x;t) + f'(x;s)$$

In the inequality I used the convexity of $f$. Don't hesitate to ask if something i unclear. Best regards

share|cite|improve this answer
    
Thank you for the solution! – Salih Ucan Feb 14 '13 at 23:50

Your Answer

 
discard

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.