Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. It's 100% free, no registration required.

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

Assume that a function $f:[0,1] \rightarrow \mathbb{R}$ is continuous. In what way condition $$\forall_{n\in \mathbb{N}} \forall_{0\leq x\leq 1-\frac{1}{n}} \exists_{0<h<1-x} \left| \frac{f(x+h)-f(x)}{h}\right|>n$$ implies nonexistence of the finite right derivative of $f$ in each point from $[0,1)$ ?

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

Consider the negation:

$$\exists_{n\in \mathbb{N}} \exists_{0\leq x\leq 1-\frac{1}{n}} \forall_{0<h<1-x} \left| \frac{f(x+h)-f(x)}{h}\right|\leq n$$

Note that the negation is equivalent to the following property. (You might want to see if you can correctly write out all the details in a proof of their equivalence.)

$$\exists_{m\in \mathbb{N}} \exists_{n\in \mathbb{N}} \exists_{0\leq x\leq 1-\frac{1}{m}} \forall_{0<h<1-x} \left| \frac{f(x+h)-f(x)}{h}\right|\leq n$$

This last property should now be recognizable as the property of having bounded right difference quotients based at some $x \in [0,1).$ The parameter $m$ allows for all possible points $x \in [0,1)$ to be included and the parameter $n$ allows for all possible (finite) bounds to be included.

Thus, the property you gave is equivalent to: At each $x \in [0,1)$ and for each right neighborhood of $x$ (where the right neighborhood belongs to the interval $[0,1]$), the function $f$ does not have bounded difference quotients where the left point is fixed at $x$ and the right point varies over points in that right neighborhood. An equivalent way of stating this is to say that, for each $x \in [0,1)$, at least one of the two right Dini derivates of $f$ at $x$ (upper right derivate or lower right derivate) is not finite. Part of the difficulty you're having might be due to the fact that this is quite a bit stronger than saying that, for each $x \in [0,1),$ the function $f$ does not have a finite right derivative at $x.$

share|cite|improve this answer
Thanks a lot for answer. – L.T Jan 10 '12 at 20:00

Fix $x\in [0,1)$. For $n$ large enough (say $\geq N(x)$, we have $0\leq x\leq 1-\frac 1n$, and we can find for $n\geq N(x)$, $0<h_n<\frac 1n$ such that $\left|\frac{f(x+h_n)-f(x)}{h_n}\right|>n$. If the limit $\lim_{h\to 0^+}\frac{f(x+h)-f(x)}h$ existed (and was equal to $l$), then we would have for $n$ large enough such that $\left|\frac{f(x+h_n)-f(x)}{h_n}-l\right|\leq 1$ $$n\leq \left|\frac{f(x+h_n)-f(x)}{h_n}\right|\leq \left|\frac{f(x+h_n)-f(x)}{h_n}-l\right|+|l|\leq l+1,$$ which is not possible.

share|cite|improve this answer
Thanks, but I dont understood why we can find $h_n$ in such a way that $h_n< \frac{1}{n}$ and difference quotient is greater then $n$. If $0\leq x \leq 1-\frac{1}{n}$ then $1-x> \frac{1}{n}$. – L.T Jan 10 '12 at 19:23

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.