3
$\begingroup$

I saw $\frac{1}{1-x}=1+x+x^2+...$ involving $\mathbb{Q}_p$ on some notes, but in calculus, it works only $|x|<1$, but in these notes, $x$ is not supposed to have this restriction. I wonder if it's a convention, in other words, if it's defined that the inverse of $(1-x)$ is defined to be $1+x+x^2+...$?

$\endgroup$
  • $\begingroup$ It is easy enough to prove that $\sum x^n$ is the inverse of $(1-x)$ in the ring of formal power series $k[[x]]$, and so for purely formal reasons, it will be the inverse in any ring where the sum actually makes sense. However, the question is: when will the sum make sense? There is a norm in $\mathbb Q_p$, and if $|x|<1$, the sum will converge. I am skeptical that it converges otherwise (sums in the p-adics converge if and only if the norm of the summands approahces 0), but I am too tired to prove this fact. $\endgroup$ – Aaron Apr 20 '17 at 3:37
  • $\begingroup$ math.stackexchange.com/questions/1829895/… $\endgroup$ – tatan Apr 20 '17 at 5:23
3
$\begingroup$

In $\mathbb{Q}_p$, a series $\sum\limits_{i\geq 0}a_i$ converges if and only if $|a_i| \rightarrow 0$ (this only hold in one direction in real analysis, for example if $a_i=\frac{1}{1+i}$ then this doesn't converge in real analysis).

So $\sum\limits_{i \geq 0} x^i$ converges if and only if $|x^i| \rightarrow 0$, if and only if $|x|<1$, which is the same criterion as for real analysis, except this time the norm is taken $p$-adically (loosely speaking, the requirement is that $p|x$).

The equality does not hold for other elements, take for example $x=1$, where the sum should be $\infty \not\in \mathbb{Q}_p$, or even $x=-1$ where the sum makes absolutely no sense at all (either in $\mathbb{R}$ or $\mathbb{Q}_p$).

$\endgroup$
3
$\begingroup$

You slightly misunderstand the nature of the statement about $\sum_{i=0}^\infty x^i$. The theorem you're used to says

If $\sum_{i=0}^\infty x^i$ converges, then it converges to $\frac{1}{1-x}$

It happens to be that, in $\mathbb{C}$, this occurs when $|x|<1$, but the theorem is provably solely off the assumption of convergence. Likewise, in $\mathbb{Q}_p$, if $\sum_{i=0}^\infty x^i$ converges, then it converges to $\frac{1}{1-x}$. In $\mathbb{Q}_p$ the criterion for that sum converging is no longer $|x|<1$. A sequence $\alpha_i\in\mathbb{Q}_p$ converges iff $|\alpha|_p\to 0$, so in $\mathbb{Q}_p$ we have:

$$\lim_{i\to\infty}|x^i|_p=0\iff \sum_{i=0}^\infty x^i=\frac{1}{1-x}$$

$\endgroup$
  • $\begingroup$ Is the equality always true in $\mathbb{Q}_p$? $\endgroup$ – Katherine Apr 20 '17 at 4:31
  • 1
    $\begingroup$ @Katherine I've edited my answer to contain the answer to that question. $\endgroup$ – Stella Biderman Apr 20 '17 at 15:11

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.