I need to prove that $f(x,y)=\arcsin \frac{x}{y}$ is continuous, but not uniformly continuous on its domain. I noticed that the domain of the function is $D_f=\{(x,y)|-y\leq x \leq y$ if $y>0$, and $y\leq x \leq -y$ if $y<0\}.$ I started to prove the continuity by the epsilon-delta deffinition, but I'm stuck at proving that $|\arcsin \frac{x}{y} - \arcsin \frac{x'}{y'}|<\epsilon$.

  • 1
    $\begingroup$ Have you tried $\sin(\arcsin(x)+\arcsin(y))$ and developing the $\sin$ and carry on your idea? $\endgroup$ – EDX May 20 at 22:36
  • $\begingroup$ @EDX I don't understand how can I relate $\arcsin \frac{x}{y}$ with $\sin (\arcsin x+ \arcsin y)$? $\endgroup$ – Emo May 21 at 9:26
  • 1
    $\begingroup$ I was meaning $\sin(\arcsin(u)+\arcsin(v))$ where you can choose $u=\dfrac{u}{v}$ and $v=\dfrac{u'}{v'}$. I could work. $\endgroup$ – EDX May 21 at 10:51
  • $\begingroup$ On that case we'll have $\arcsin \frac{x}{y} - \arcsin \frac{x'}{y'} = \arcsin \left( \frac{x}{y} \sqrt{1-\frac{x'}{y'} - \frac{x'}{y'} \sqrt{1-\frac{x}{y}} \right)$. Now how should we prove that the RH in absolute values is small enough when points are sufficiently close to each other? $\endgroup$ – Emo May 22 at 10:20
  • 2
    $\begingroup$ Why can't you just use the fact that both arcsin and $x/y$ are continuous? $\endgroup$ – Sam May 22 at 22:52

To show that the function $f$ is not unifromly continous first recall that $\operatorname{arcsin} 1 =\tfrac{\pi}2$ and $\operatorname{arcsin} \tfrac 12 =\tfrac{\pi}6$. Thus for each natural $n$ we have $f\left(\tfrac 1n, \tfrac 1n\right)= \tfrac{\pi}2$ and $f\left(\tfrac 1n, \tfrac 2n\right)= \tfrac{\pi}6$, whereas $\left|\left(\tfrac 1n, \tfrac 1n\right)- \left(\tfrac 1n, \tfrac 2n\right)\right|=\tfrac 1n$.

| cite | improve this answer | |

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.