Thread: prove limit by definition

prove that by epsilon delta
defintion.
we need to prove that their is so for every
their is x and
so we are free to choose our epsilon and our x
i choose
x cannot be equal to -1 because of the denominator.
and we need to choose x for which
-> ->
-> ->
but x=-1 there so we have a problem.
where is the problem in my way???


the prof solved it by

cant understand why he choose such x

Originally Posted by transgalactic

prove that by epsilon delta

defintion.
we need to prove that their is so for every
their is x and
so we are free to choose our epsilon and our x
i choose
x cannot be equal to -1 because of the denominator.
and we need to choose x for which
-> ->
-> ->
but x=-1 there so we have a problem.
where is the problem in my way???

the prof solved it by

cant understand why he choose such x

Since the actual limit is 2/3, you will always be able to show that for x close enough to 2 that |x/(x+1)-1|>0.3, so you need to choose epsilon greater than zero but less than 0.3 (well actually any thing less than 1/3 will do). Then show that for all delta>0 there is always some x such that |x-1|<delta and |x/(x+1) - 1|>epsilon.

CB

what is the problem of my way.
??

Consider <_1. Then if x satisfy's 0< |x-2|< , F(x)<3/4. Hence |F(x)-1|>1/4. Choosing <_1/4 completes the proof.

i cant because its for every delta

and can you tell me where am i wrong
?

For any delta>1, x values in (1,3) will still have to satisfy |F(x)-1|< epsilon.

It's not true that you are totally free to choose any ε that you like.

As CaptainBlack told you, the actual limit is 2/3.

|1 - 2/3| = 1/3. So, if you use an ε ≥ 1/3, it is possible to find a δ > 0 such that for all x for which 0 < | x - 2 | < δ it follows that | x/(x+1) - 1 | < ε .

That's why CB said you need to choose ε < 1/3 .