4

In the book Roberts, Varberg, Convex functions, on pp.9-10 is proved that

If $f:(a,b)\rightarrow \mathbb R$ is continuous and convex then $f$ is absolutely continuous on each $[c,d]\subset (a,b)$.

How can this be extended to the case when the domain of $f$ is instead $[a,b]$?

Edit

I wish to prove the following theorem:

If $f:[a,b] \rightarrow \mathbb R$ is continuous and convex then $f(b)-f(a)=\int_a^b f'(t)dt$.

Thanks

Richard
  • 4,564

2 Answers2

3

Suppose that $f\colon (a,b)\to \mathbb{R}$ is convex. Consider the function

$$F(x,t) = \frac{f(x+t) - f(x)}{t}\quad(t\neq 0).$$

Then $F$ is increasing in each variable. So $F$ is uniformly bounded on each compact subinterval $[c,d]\subset (a,b)$. In particular, $f$ is Lipschitz on $[c,d]$ hence absolutely continuous.

Tomasz Kania
  • 16,996
2

Hint. It should be enough to use the case from the book, together with:

If $\phi$ is nondecreasing, then $$ \int_a^b \phi(t)\,dt = \lim_{\alpha\searrow a, \beta \nearrow b} \int_\alpha^\beta \phi(t),dt $$

GEdgar
  • 117,296
  • Does there always exist integral $\int_a^bf'(t)dt$ for convex $f$ ? In general the values $f'(a)$, $f'(b)$ can be in finite for convex continuous $f$. – Richard Jan 05 '15 at 13:50
  • 1
    But the limit of $f(\beta) - f(\alpha)$ is finite, right? – GEdgar Jan 05 '15 at 13:55
  • Ok, this limit is $f(b)-f(a)$, what is needed. Many thanks for explanation. – Richard Jan 05 '15 at 14:03