0
$\begingroup$

I know a couple of different ways to prove a limit like the $\epsilon-\delta$ approach and squeeze theorem. I know you can find a limit using squeeze theorem, definition of derivative, l'hospital, taylor series, and some algebraic or trigonometric approach.I know finding a limit algebraically does not count as a proof...However some seem to think finding a limit the l'hospital way is proof. If so why? Why is l'hospital proof while the algebraic way is not? Also please add any other ways to my list above to prove limits if you can think of any other general ways. Thanks kindly.

$\endgroup$
  • 2
    $\begingroup$ I disagree with the premise "I know finding a limit algebraically does not count as a proof". $\endgroup$ – Git Gud Nov 10 '15 at 20:50
  • 2
    $\begingroup$ Since all $=$ signs are perfectly justified here, this can be seen as a proof. $\endgroup$ – Alexander Nov 10 '15 at 20:53
  • 3
    $\begingroup$ @randomgirl The concept of "proof" means different things to different people in different contexts. After having proved that the "limit function" is additive and multiplicative, I see no reason why that sequence of equalities shouldn't be considered a proof (in a not too strict sense of the word). $\endgroup$ – Git Gud Nov 10 '15 at 20:54
  • 1
    $\begingroup$ They don't always say that. $\endgroup$ – The Chaz 2.0 Nov 10 '15 at 20:54
  • 2
    $\begingroup$ @randomgirl Because direct substitution depends on the fact that $x\mapsto \frac 1 x$ is continuous and that might require a $\varepsilon$-$\delta$ proof. The thing with your example is that it could very well be asked after continuity of $x\mapsto x+2$ is proved and after some other well known facts proved. In this case it depends on context. $\endgroup$ – Git Gud Nov 10 '15 at 21:34
0
$\begingroup$

It depends what you mean. An $\epsilon-\delta$ proof follows directly from the formal definition of a limit, however evaluating limits in other means obviously works and is much easier than using an $\epsilon-\delta$ argument every time. These methods have been proved to work so they are valid ways of "proving" limits however in a real-analysis class they usually want you to prove everything from the axioms, so for example you can prove a limit using L'Hopital's rule but you'd have to prove L'Hopital's rule first, in calculus however they just teach you the evaluation methods and you have to just assume that they work. Hope this helps!

| cite | improve this answer | |
$\endgroup$

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.