I am trying to solve exercise 9.4.4 from Tao's Analysis I. It says:

Prove the following theorem:

Let $p$ be a real number. Then the function $f : (0, \infty) \to \mathbb{R}$ defined by $f(x) := x^p$ is continuous.

Hint: Prove that for all real numbers $p$ $\lim_{x \to 1} x^p = 1$. Then apply some basic facts about exponentiation.

The basic facts he refers to (he refers to the theorem by number) are:

Let $x,y>0, q,r \in \mathbb{R}$. Then:

  1. $x^q > 0$
  2. $x^{q+r} = x^q x^r$
  3. $(x^q)^r = x^{qr}$
  4. $x^{-q} = 1 / x^q$
  5. if $q>0$, then $x>y$ iff $x^q>y^q$
  6. If $x>1$, then $x^q>x^r$ iff $q>r$. If $x<1$, then $x^q>x^r$ iff $q<r$.

BTW the textbook defines exponentiation $x^\alpha$ for real $\alpha$ to be $\lim_{n \to \infty} x^{q_n}$, where $(q_n)_{n=1}^\infty$ is any sequence of rational numbers converging to $\alpha$.

I think I can prove it using other means - first proving that for any integer $n$ $f(x) = x^n$ is continuous, then prove that for any positive integer $n$ $f(x) = x^{1/n}$ is continuous, then prove that for any rational $p$ $f(x) = x^p$ is continuous, then prove continuity for real $p$.

But I have no idea how to do it using the author's hint. I have proved that $\lim_{x \to 1} x^p = 1$, but I've got no idea what to do with this fact. Please help me figure it out.


You want to show that, for every $a>0$, $$ \lim_{x\to a} f(x) = f(a) $$ i.e., $$ \lim_{x\to a} x^p = a^p $$ and you want to infer this from the case $a=1$. Observing that ${x^p}/{a^p} = (x/a)^p$, can you prove that $$ \lim_{x\to a} \frac{x^p}{a^p} = 1 $$ and conclude from there?

  • $\begingroup$ This helped me, I have not solved it, and the steps you suggested turned out to be very simple. I wonder why I missed this and what general technique of proving things I should learn to see such proofs. Thank you. $\endgroup$ – CrabMan Sep 27 '18 at 10:15
  • $\begingroup$ @CrabMan I am not aware of a general technique, but in general trying to reverse-engineer the problem helps (to start a proof; the proof itself should not go backwards). So, on rough paper, start with the conclusion, and try and poke/massage (no need to be too rigorous) it until you get something that resembles your assumptions. Then, once you have an idea of a path to go the other way around, you can formalize the whole thing and fill the gaps. $\endgroup$ – Clement C. Sep 27 '18 at 19:19
  • $\begingroup$ Btw I've asked Terry Tao by email, he replied that indeed this property is supposed to be included in the list, and it is missing by mistake. $\endgroup$ – CrabMan Sep 28 '18 at 6:16

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.