Take the 2-minute tour ×
Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. It's 100% free, no registration required.

I need to prove the following limit using definition only:
$$\lim_{x\to -\infty} \frac{(7x+3)}{(x-1)} =7.$$

The definition is: for any $\epsilon >0$. there is a $\delta$ so $x<\delta \implies |f(x)-L|<\epsilon$

In order to show that $|f(x) - L| < \epsilon$ I assumed $\delta =1, \delta=0$ and in the end i showed $\delta= 10/\epsilon$. The problem is I don't know which one to choose: $\min\{\delta_1, \delta_2, \cdots\}$ or $\max\{\delta_1, \delta_2, \cdots\}$. In the normal definition of a limit i know that we always choose the minimum delta but i am not sure what do here.

My second question is: can I assume many things about $\delta$ then just choose minimum/maximum, is it ok?

Thanks for help.

share|improve this question
Are you using $B$ to represent $\delta$ = \delta? –  amWhy Nov 8 '13 at 14:58
Yes, the definition is: there is a B so x<B => |f(x)-L|<E –  Michael Nov 8 '13 at 15:00
Your definition is logically wrong. "For any $\epsilon>0$" always comes at the beginning, never at the end. Moreover, it is not legitimate to assume $\delta=0$, since you want $\delta>0$. However, you can assume that $\delta$ is "small enough", and finally select the minimum value of $\delta$. –  Siminore Nov 8 '13 at 15:07
$\epsilon$ = \epsilon, and $\delta$ = \delta –  amWhy Nov 8 '13 at 15:07
@Siminore I edited the post, epsilon is at the begining now. But $\delta$ does not have to be > 0. But you can assume it. –  Michael Nov 8 '13 at 15:12

1 Answer 1

up vote 2 down vote accepted

What we want in the definition is

For any $\epsilon > 0$, there is a $\delta$ such that $x \lt \delta \implies |f(x) - L| \lt \epsilon.$

(Definition since corrected in the OP).

For any $\epsilon > 0$, we can make $\delta$ as small as we'd like to ensure the implication holds, and in the end, we select the minimum $\delta$.

share|improve this answer
$\delta$ doesn't have to be >0 but you can assume it. There are people who add it to the definition and others don't. Thanks for answering, could you please answer my second question? –  Michael Nov 8 '13 at 15:14
It depends what you mean when you say "assume many things." Often times, we can design each $\delta$ to be a function of $\epsilon$ so that for any $\epsilon >0$, we can ensure the existence of a $\delta$ that makes the implication true. –  amWhy Nov 8 '13 at 15:21

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.