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

$f\in C[0,2\pi]$ and $f_\delta := f$ in $[\delta , 2\pi - \delta],f_{\delta}(0)=f_{\delta}(2\pi)=\frac{1}{2}(f(0)+f(2\pi))$ and $f_\delta$ is linear in $[0,\delta]$ and also in $[2\pi - \delta , 2\pi]$

For every $\epsilon > 0$ there exists a $\delta > 0 $ such that : $||f-f_\delta||<\epsilon$.

How to prove this?

share|cite|improve this question
Which norm are talking about? – Norbert Feb 2 '13 at 12:13
the norm induced by : $||f||:= (f,f)^{1/2} = (\int_{0}^{2\pi}f^2)^{1/2})$ – bakabakabaka Feb 2 '13 at 12:15
I think this question raised when you were thinking about Fourier series, but as the question stated it is not related to Fourier series at all – Norbert Feb 2 '13 at 12:17
I tried to correct the title, is it better suited now? – bakabakabaka Feb 2 '13 at 12:22
up vote 1 down vote accepted

Since $f=f_\delta$ on $[\delta,2\pi-\delta]$, then $$ \Vert f-f_\delta\Vert= \left(\int\limits_{[0,\delta]}|f(x)-f_\delta(x)|^2 dx+\int\limits_{[2\pi-\delta,2\pi]}|f(x)-f_\delta(x)|^2 dx\right)^{1/2} $$ Let $M=\max\limits_{x\in[0,2\pi]}|f(x)|$, then for all $x\in[0,2\pi]$ we have $|f_\delta(x)|\leq M$ and we conculde $$ \Vert f-f_\delta\Vert\leq\left(\int\limits_{[0,\delta]}(2M)^2 dx+\int\limits_{[2\pi-\delta,2\pi]}(2M)^2 dx\right)^{1/2}=2M(2\delta)^{1/2} $$ So for a given $\varepsilon$ it is enough to set $\delta=\varepsilon^2/(8M^2)$.

share|cite|improve this answer
Thank you very much! – bakabakabaka Feb 2 '13 at 14:30

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.