Limit problem

Calculus Level 2

From the graph of f f , evaluate

lim x 6 f ( f ( x ) ) . \Large \lim_{x\to6}f(f(x)).

5 6 7 Does not exist

This section requires Javascript.
You are seeing this because something didn't load right. We suggest you, (a) try refreshing the page, (b) enabling javascript if it is disabled on your browser and, finally, (c) loading the non-javascript version of this page . We're sorry about the hassle.

4 solutions

Micah Wood
Dec 8, 2014

Observe that when x x is close to 6 6 but smaller ( 5.999 5.999 ), we can see that f ( x ) f(x) is close to 2 2 , but bigger.

And when x x is close to 6 6 but bigger ( 6.001 6.001 ), we can see that f ( x ) f(x) is close to 2 2 , but bigger.

So we can see that whenever x x approaches 6 6 , in both sides f ( x ) f(x) approaches y y -value of 2 2 from above.

So lim x 6 f ( f ( x ) ) = lim x 2 + f ( x ) = 5 \displaystyle \lim_{x\to6}f(f(x)) = \lim_{x\to2^+}f(x) = \boxed{5}

very good limit problem.

Akash Mandal - 6 years, 5 months ago

Log in to reply

i don't think the problem is right. there aren't a sign + on the limit target 6 so we can't say that the limit of f(f(x)) for x to 6 is f(f(x) for x to 2+, why not (f(x) for x to 2-??? in my opinion the limit does not exist for x to 2, The only limit that are possible are x to 2+ and x to 2- . This two limit are different so limit for f(x) for x to 2 noes not exsist!

Roberto Lo Prejato - 5 years, 7 months ago

Log in to reply

Notice that when x x approaches 6, the y y value of the graph approaches 2 from above, not from below. Therefore, it is safe to say that

lim x 6 f ( f ( x ) ) = lim x 2 + f ( x ) \displaystyle \lim_{x\to6}f(f(x)) = \lim_{x\to2^+}f(x)

Andrew Ellinor - 5 years, 7 months ago

Log in to reply

@Andrew Ellinor don't we have that f f has a limit at 6 iff ϵ > 0 δ > 0 \forall \epsilon>0 \exists \delta>0 such that if 0 < x 6 < δ , x R , 0<|x-6|<\delta, x\in R, then f ( x ) 2 < ϵ |f(x)-2|<\epsilon ? I ask because this means that 6 δ < x < δ + 6 6-\delta<x<\delta +6 . I don't know what the formula for f f is but, for example, , 5.999 = 6 . 001 , 5.999=6-.001 seems to be in the domain and 6 0.1 < 6 0.001 < 6 < 6 + 0.1 6-0.1<6-0.001<6<6+0.1 . At least it fits the definition right? So I don't know why the y y value approaches 2 only from above, or at least I don't know what that means. Certainly, it has both a right-hand and a left-hand limit.

Jorge Alonso - 5 years, 5 months ago

Log in to reply

@Jorge Alonso Yes, and those limits are the same. As x -> 6, f(f(x)) never gets near 7.

Take a small deleted neighborhood of 6: ( 6 δ 0 , 6 ) ( 6 , 6 + δ 0 ) (6 - \delta_{0}, 6)\cup(6, 6 + \delta_{0}) . The image never drops below 2, nor equals 2, so the image is something like ( 2 , 2 + δ 1 ) (2, 2 + \delta_{1}) . Take the image of that under f, and you get something like ( 5 δ 2 , 5 ) (5 - \delta_{2}, 5) . True?

Whitney Clark - 5 years, 5 months ago

Log in to reply

@Whitney Clark I think I see your point. What you're saying is that because lim x 6 f ( x ) = 2 \lim_{x\rightarrow 6}f(x)=2 and because f ( x ) f(x) are now the inputs, we are only allowed to use values not smaller than 2. So f ( x ) 2 f(x)\geq 2 as x 6 x\rightarrow 6 . Then for x 6 , lim y 2 f ( y ) = lim y 2 + f ( y ) , y 2 x\rightarrow 6, \lim_{y\rightarrow 2}f(y)=\lim_{y\rightarrow 2^{+}}f(y), y\geq 2 and y = f ( x ) y=f(x) . I agree with this.

Jorge Alonso - 5 years, 5 months ago

@Andrew Ellinor So if say, we had an upside down parabola-type segment of a function as opposed to the right side up parabola-type function, then x approaching from the bottom, then you would look at the left side of the function and conclude the limit is 7?

Oli Hohman - 5 years, 5 months ago

@Andrew Ellinor No.. the value of f ( x ) f(x) at x = 6 x=6 is 2 \boxed{2} .

Rodolfo Marinho - 5 years, 7 months ago

Log in to reply

@Rodolfo Marinho Yes, and note that while x approaches to 6 6 , f ( x ) f(x) approaches 2 2 FROM ABOVE; in interval ( 2 , 10 ) (2,10) , there is no x x such that f ( x ) < 2 f(x)<2 ...

Micah Wood - 5 years, 7 months ago

@Rodolfo Marinho I think the graph is not supposed to go through the point (6 , 2) but slightly above it. That's why it's the limit when x is 6 is slightly above 2 that's why it would be the limit when x is lightly above 2 which would be 5

James Waychoff - 5 years, 5 months ago

@Rodolfo Marinho What do you mean? It was asking about limits, not values.

Whitney Clark - 5 years, 5 months ago

I agree with you. The overall limit as x -> 2 does not exist. To get the answer they got it would have to specify the lim of f(x) as x->2+.

Matthew Wethington - 5 years, 7 months ago

Log in to reply

@Matthew Wethington And it did. As x -> 6, f(x) -> 2+ only, and not to 2-.

Whitney Clark - 5 years, 5 months ago

Let f(x) = y. The limit of y as x goes to 6 is 2. Especially around that point, the limit of f(y) as y goes to 2 is always 5, as you can't approach y to 2 from below (notice that at the value of x around 6, all the possible values of y is above 2)

Gian Sanjaya - 5 years, 5 months ago

I agree, there is not a + in the original saying the limit is approaching from the right. So "does not exist" is the correct answer, since it is just the limit.

David Lauinger - 5 years, 7 months ago

Log in to reply

@David Lauinger Why doesn't it exist? f ( x ) f(x) approaches to 2 from upward when x approaches to 6 from either right or left. So It is true that lim x 6 f ( f ( x ) ) = lim x 2 + f ( x ) \displaystyle \lim_{x\to6}f(f(x)) = \lim_{x\to2^+} f(x) .

I've simulated graph .

Micah Wood - 5 years, 7 months ago

Log in to reply

@Micah Wood The limit at 2 doesn't exist because it doesn't approach a single number. from the left it approaches 7, and from the right it approaches 5.

David Lauinger - 5 years, 7 months ago

Log in to reply

@David Lauinger Original limit is equal to limit at 2 from right side...

Micah Wood - 5 years, 7 months ago

Log in to reply

@Micah Wood In the limit of a composite function you only calculate the limit once (either for the inner or for the outer function, not for both).

Rodolfo Marinho - 5 years, 7 months ago

Log in to reply

@Rodolfo Marinho Excellent point, and a correct one!

a bagchi - 4 years, 11 months ago

@David Lauinger But it doesn't approach from the left, but from the right only, since f ( x ) 6 f(x) \geq 6 .

Whitney Clark - 5 years, 5 months ago

@David Lauinger It's not the limit at 2, but the limit at 2+, since the limit at 6 is 2, through greater values ONLY.

Whitney Clark - 5 years, 5 months ago

totally agree

Alireza Rasekh - 5 years, 4 months ago

I thought this was an excellent problem. I'm surprised that anyone is finding difficulty in agreeing with Micah's solution.

Sheridan Halls - 5 years, 6 months ago

Log in to reply

That's because this problem requires thinking and not just blind application of rules.

Whitney Clark - 4 years, 4 months ago

I don't understand how people don't get this. Near x=6, f(x) is always most definitely going to be greater than 2, so x=2 must always be approached from the the right side... no matter what direction you approach x=6... Great problem +1

Bryan Hung - 5 years, 5 months ago

Log in to reply

People don't get it because they don't think but blindly follow rules.

Whitney Clark - 4 years, 4 months ago

Log in to reply

The government is putting fluoride in the water to make us follow made up rules. Wake up sheeples!

Julia Fridlund - 2 years, 5 months ago

People (including me) probably got confused by the open and closed circles in the given graph (,don't ask why). Actually, I've just understood the solution through your comment (, thanks).

It's a nice and tricky problem! :)

チャレス アギアル - 5 years, 6 months ago

Log in to reply

tricky yes, nice - not :D

check my comment on the domain of f(x) and fof(x)

Iacob Juc - 5 years, 5 months ago

At point x=6

f(x) is continuous function with f(6) = 2 and

Lim[x-->6+] f(x) = Lim[x-->6-]f(x) = 2+

then Lim f(f(x)) when x-->6 is

f(2+) --> 5

Answer 5.

Sladjan Stankovik - 5 years, 5 months ago

This is a brilliant problem. But I want to ask isn't the lim x->6 going to disappear when we already found the f(x)? There is no longer a 'x' term there.

Kwg Dennis - 5 years, 5 months ago

Log in to reply

But we are not looking for f(x); we are looking for the limit. As x -> 6, f(x) -> 2+, so that f(f(x)) -> 5-.

Whitney Clark - 5 years, 5 months ago

Log in to reply

Firstly, x approaching 6 does not make f(x)--> 2+, it makes it 2. Just for the shake of arguments, even if that 2+ assertion was somehow correct, why then we change the limit problem with x approaching 2+? Didn't we just claim f(x), not x , that is 2+ ? Also, nowhere in the limit theory one can change the approaching value for the independent variable (read x) unless we are using an explicit substitution. On the other hand, limit theory of composite functions (any textbook will show that) will tell anyone that the current limit reduces to f(2), which is defined in the graph as 6. Please see my explanation in above for that. This is a nice problem, and the extended discussion on this thread underscores how much attention is provoked in this community by this problem. So, at the end, I thank the original poster. I just do not agree with the final answer provided.

a bagchi - 4 years, 11 months ago

Log in to reply

@A Bagchi Except it was not asking for f(lim f(x)), but lim f(f(x)). Limits are never concerned with the value of a function, but only with its behavior approaching certain values.

If all else fails, go back to the original definition of limit. To get f(f(x)) within a small error of 5 (which will end up happening through lesser values only), it suffices to get f(x) within a small difference of 2, through greater values ONLY. And to get f(x) within a small error of 2, through greater values only, it suffices to get x within a small difference of 6. Thus, as x -> 6, f(x) -> 2+, and f(f(x)) -> 5-.

Did I make any mistakes? Remember, theory can be flawed, so when it comes down to it, stick with what you know.

Whitney Clark - 4 years, 4 months ago

You are correct Dennis, see my explanation for 6.

a bagchi - 4 years, 11 months ago

I answered with the left neighbour, so I got 7, it says that its wrong, why ?

John Arthur Samuel - 5 years, 3 months ago

Log in to reply

I usually like to think of it was what happens to the behavior of the whole as x goes to whatever. Like in x^2 + 7, we have, as x -> 3: x^2 -> 9, and x^2 + 7 -> 16. So the limit is 16.

In this case, as x goes to 6 from BOTH sides (x -> 6), f(x) goes to 2 through greater values ONLY (f(x) -> 2+). And at the same time, f(f(x)) goes to 5 through LESSER values only (f(f(x)) -> 5-), just like as x goes to 2 through greater values, f(x) goes to 5 through lesser values.

If it helps, go back to the definition of limit.

Whitney Clark - 5 years, 3 months ago

For those that are still in disagreement, we can ask the same question for lim[x->0] f(f(x)), where f(x) = 1/x. The limit here is clearly 0, even though the limit of 1/x doesn't exist at 0.

Whitney Clark - 4 years, 4 months ago

Wow amazed by this problem

Shubham Kumar - 3 years, 1 month ago

The limit does not exist For the constant sequence xn = 6 we have f(f(xn)) = f(f(6)) = f(2) = 6. But for the sequence yn :6.1, 6.01, 6.001, ... f(f(yn)) tends to 5.

Jacques Labelle - 2 years, 11 months ago

Log in to reply

No, x is not supposed to ever equal 6, it's only supposed to approach it. That it does something different at exactly 6 is unimportant.

Whitney Clark - 2 years, 11 months ago

This limit problem is not an accurate problem.

Matthew Wethington - 5 years, 7 months ago

Log in to reply

Notice that when x x approaches 6, the y y value of the graph approaches 2 from above, not from below. Therefore, it is safe to say that

lim x 6 f ( f ( x ) ) = lim x 2 + f ( x ) \displaystyle \lim_{x\to6}f(f(x)) = \lim_{x\to2^+}f(x)

Andrew Ellinor - 5 years, 7 months ago

Log in to reply

Safety is not a mathematical argument. However, correctness is. What is exactly meant by this above statement of yours? Can you provide anyone of us a solid reference to any book (or paper) where you may have seen a similar equation? Thanks.

a bagchi - 4 years, 11 months ago

No, it's not safe to say that.. There is no such thing as saying that f ( x ) = 2 + f(x) = 2^+ !

Rodolfo Marinho - 5 years, 7 months ago

Log in to reply

@Rodolfo Marinho It's true that f ( 6 ) = 2 f(6) = 2 , but it is also true that lim x 6 f ( f ( x ) ) = lim x 2 + f ( x ) \displaystyle \lim_{x\to6}f(f(x)) = \lim_{x\to2^+}f(x) .

I agree that there is no such thing as saying that f ( x ) = 2 + f(x) = 2^+ . However, that is not what the statement lim x 6 f ( f ( x ) ) = lim x 2 + f ( x ) \displaystyle \lim_{x\to6}f(f(x)) = \lim_{x\to2^+}f(x) implies.

Andrew Ellinor - 5 years, 7 months ago

Log in to reply

@Andrew Ellinor Composition limit laws state... If f is continuous at b and the limit as x goes to a of g(x) = b, then

the limit as x goes to a of f(g(x)) = f(b) = f(the limit as x goes to a of g(x)).

The restrictions that are placed on the values of f(x) do not create a situation whereby the limits approach something. The limit at f(2) does not exist.

If we use the composition law it states the limit would in fact be 6. But it isn't continuous at b.

There is a reason there is so much contest around this question because unless it states explicitly that we are looking for a one-sided limit, the question is flawed.

There is a one-sided limit in the initial problem so there is nothing special about the question, in fact, the initial problem is ambiguous.

The report button doesn't work or I would have reported it again.

For the author Micah Wood, having a staff agreement doesn't make you correct. This forum isn't just for solutions. For respectful non-insulting posts (which is everything I've seen including your posts but seeing you defensive is counterproductive) The forum isn't for solutions it is to "DISCUSS" solutions. You need to not be so defensive. Please feel free to discuss my solution even if you disagree. Please use well formed arguements with rules (not pictures). We all can visualize your logic. Please provide a rule, theorum or argument that proves your interpretation.

In my opinion, could be seen to not exist as a complete limit. As a one-sided limit yes, but not as a complete limit.

Eric Belrose - 5 years, 5 months ago

Log in to reply

@Eric Belrose The laws have exceptions, just like how if ax = bx, then you DON'T have a = b if perhaps x = 0. In this case, you appeal to the definition.

Whitney Clark - 5 years, 5 months ago

@Rodolfo Marinho Who said anything about f ( x ) = 2 + f(x) = 2^+ ? There is difference between lim x 6 f ( f ( x ) ) = lim x 2 + f ( x ) \lim_{x\to6}f(f(x)) = \lim_{x\to2^+}f(x) and f ( x ) = 2 as x 6 f ( x ) = 2 + f(x)=2 \text{ as }x\to6\Longrightarrow f(x)=2^+ or whatever you are thinking.

First is certainly possible while second is nonsense.

Micah Wood - 5 years, 7 months ago

@Rodolfo Marinho You can't use the limit's theorem for composition of fuctions, because f f isn't continuos on 2 2 . So I agree with Micah and Andrew, or what are you saying that?

Angel T. López - 5 years, 5 months ago

@Rodolfo Marinho That's not what it's saying at all. As x->6, f(x)->2+ (i.e. through greater values ONLY), so f(f(x)) -> 5.

Whitney Clark - 5 years, 5 months ago

@Rodolfo Marinho That's not what the question was asking, though. It wasn't asking about f(x), but only its limit.

Whitney Clark - 5 years, 3 months ago

How so? Do I need to specify that f ( x ) f(x) contains point ( 6 , 2 ) (6,2) ?

Micah Wood - 5 years, 7 months ago

Log in to reply

No. Exact values aren't important in limits.

Whitney Clark - 5 years, 5 months ago

Using the limit definition of differentiation, in order for a function to be differentiable, it must continuous. As F(x) is not differentiable at x=2, therefore the function is NOT continuous at x=2. As it is not continuous at x=2, therefore the limit does NOT exist at x=2.

As x->6, we can see that f(f(x)) approaches 2 (it does not state which side it approaches from, therefore it will approach 2 from both sides as well). And since in the above statement, no limit exists for x=2, therefore the limit DOES NOT EXIST.

Colin Zhao - 5 years, 5 months ago

Log in to reply

Differentiable implies continuous, not the other way around. f(x) = |x| is continuous at x=0 but it isn't differentiable there.

A Former Brilliant Member - 5 years, 5 months ago

Log in to reply

f(x)=|x| is certainly differentiable at x=0. The slope of this function at x=0 is 0 and is also equal to df/dx.

Shyam Sundar - 5 years, 5 months ago

Log in to reply

"it does not state which side it approaches from, therefore it will approach 2 from both sides as well"

Nice logic. It's like saying "My friend doesn't tell me if he wants a drink, therefore he must wants drink."

Micah Wood - 5 years, 5 months ago

You can't just replace the limit approaching 6 (no sign) to approaching 2+. The positive means from the right, not from above. Even though 2 is a minimum it doesn't give a sign to the limit just because there is nothing below that indicated value.

Dany Yaacoub - 5 years, 5 months ago

Log in to reply

Ok report my problem... I am waiting.

Try sketch f ( f ( x ) ) f(f(x)) using the graph of f f at around x = 6 x=6 . It should give you aha moment.

Micah Wood - 5 years, 5 months ago

"From the right" and "from above" mean the same thing, only except for the way the number line is drawn. The symbol 2+ means we are taking limits at 2, through greater values only.

Or, take a near-infinitesimal interval containing 6; then remove 6 from it. Apply f - or, take its image under f. You'll get (2, 2 + i), for some infinitesimal i. Apply f again, and you see you are only dealing with inputs greater than 2 - not 2 itself, nor anything less.

Whitney Clark - 5 years, 5 months ago

but f(x) and f(f(x)) do not have the same domain of definition, since f(x) maps R to a subdomain, i.e. f:R->[2, +inf); not sure about that, could also be (2, +inf), an open interval on 2.

so the composite function fof is either fof: (2, + inf) -> (2, +inf) or [2, +inf) -> [2, +inf); given these two formulations, the limit could be both 5 and 6;

note that in the case where fof is defined over [2,+inf), the limit is 6!, since the function is well defined in 2 as being 6; it is not continuous, but defined!

Iacob Juc - 5 years, 5 months ago

Sorry but i don´t agree with your resolution. If f is continuos in a vicinity of 6 then the problem is equivalent to f(lim(x->6)(f(x))=f(2)=6.

Anabela Matoso - 5 years, 3 months ago

Log in to reply

How do you reason? How do you get that from the definition and/or theorems?

Whitney Clark - 4 years, 3 months ago

Log in to reply

That is the definition of the limit of composite functions.

Shravan Balaji - 2 months ago

I do dissent with your solution is a fact and all of us agree that for f(6) = 2 . We also see by the graphic that precisely for x=2 , f(2) has three values then we cannot say that value f(2) is 5. I challenge how you cannot say is not 6 otherwise it will not had any sense this point in the graphic.

Mariano PerezdelaCruz - 5 years, 6 months ago

Log in to reply

You do realize that we are dealing with limit, and not f ( f ( 6 ) ) f( f(6) ) , right...?

Moreover f ( 2 ) f(2) doesn't have three values.

Micah Wood - 5 years, 6 months ago

f(f(x)) is a discontinuous at 2. No limit exists.

Continuous Function

There are three conditions that need to be met by a function f in order to be continuous at a number a. These are:


  f(a) is defined.
 Limx→af(x) exists.
 Limx→af(x)=f(a)

Discontinuous Function

If any one or more of these three conditions is not true for f at "a", then the function f is called a discontinuous function at "a".

Jay Leavitt - 5 years, 5 months ago

Log in to reply

So... why does limit must not exist?

Micah Wood - 5 years, 5 months ago

DNE Both right and left limits do not converge to the same value.

Carl Olimb - 5 years, 5 months ago

Log in to reply

"Both the left and right hand limits as x x approaches 6 both approach 2 from the right! This is because the left and right approaches to x = 6 x=6 come at the y y -value of 2 from above, not from below."

Micah Wood - 5 years, 5 months ago

the limit of f(x) does not exist since when we approach 6 from the left and right, the limits are different so they don't exist. therefore, we cannot say that limf(f(x)) exists.

Eugenia Par - 5 years, 5 months ago

Log in to reply

Wrong. As x approaches 6 from the left and the right, f(x) approaches 2 from the right only - or, to be more accurate, as x approaches 6 through greater and lesser values, f(x) approaches 2 through greater values only. Therefore, f(f(x)) approaches 5 only, through lesser values.

Whitney Clark - 5 years, 5 months ago

I cannot agree with the suggested answer here. This is the limit of a composite function f(f(x)), which by definition is equal to f(lim f(x)), as x approaches 6, provided that 'inner' limit exists. And it does, as we can graphically deduce from the enclosed figure, to be equal to 2. So the original limit problem reduces to equal f(2) = 6, and not 5.

a bagchi - 4 years, 11 months ago

Log in to reply

That is not the definition. The definition is, lim_(x->x[0]) f(x) = L iff for every epsilon > 0 there exists delta > 0 such that for all x, 0 < |x - x[0]| < delta imples |f(x) - L| < epsilon. (Here, x[0] is used for x-naught. I'm not good with LaTeX.) Taking shortcuts in unusual situations can lead to problems.

Whitney Clark - 4 years, 4 months ago

The problem is that for a limit to exist it must exist on both sides. So to state that the problem is when the RH limit only still doesn't mean that the limit exists, even if the problem only approaches from 2+. This is not the same as the f(x) not being f(a) or this approaching from the right. If the RH limit doesn't match the LH limit then the limit doesn't exist that is basic calculus! No complaint just math!

Eric Belrose - 5 years, 5 months ago

Log in to reply

So why didn't you report my problem?

Can you try and sketch f ( f ( x ) ) f(f(x)) at around x = 6 x=6 using given graph of f f ? When you do, you will see that f ( f ( x ) ) f(f(x)) has "parabolic" curve opening downward with "vertex" at x = 6 x=6 like this .

Hope this clears things up.

Micah Wood - 5 years, 5 months ago

Log in to reply

The report function doesn't work. I understand your logic, but I don't agree. One-sided limits are not a limit unless they are equal.

In your logic there doesn't even seem to be a one-sided limit.

I don't believe that you can ignore the fact that the limit doesn't exist in f around the value despite the fact that x is 6.

I would love to see a clean epsilon delta, or a limsup/liminf proof of the limit. There might be one but there is so much repetition. But I totally understand your logic.

By arguing with you no one is attacking you. You should edit your post to say keep it respectful these forums are to discuss solutions, they aren't just to post solutions but to discuss them. You don't need to be childish and state go ahead report me, but I understand how you feel attacked. I have been wrong a million times I have no issue with that, I learned long ago to accept it. It comes with time. I don't see conclusive proof in anything that has been posted thus far. I see a logic, but not explicit proof, or a rule.

The initial question is ambiguous. If you say that there is a RH limit then the answer DNE is correct by your own logic.

Have a great day, and don't feel bad because math is complex and requires a great deal of discussion, definition and limitation to maintain precision. The fact that you started a discussion is great.

Eric Belrose - 5 years, 5 months ago

Log in to reply

@Eric Belrose "I would love to see a clean epsilon delta, or a limsup/liminf proof of the limit." First no such thing as "clean proof" here, as there is only graph to use, no function. Secondly since you stated that report system is "broken," here's copy-paste of epsilon-delta rough proof:

"To say that lim x 6 f ( f ( x ) ) = 5 \displaystyle \lim_{x\to6}f(f(x))=5 is to say that ϵ > 0 , δ \forall\epsilon>0, \exists\delta , such that x \forall x if x 6 < δ |x-6|<\delta , then f ( f ( x ) ) 5 < ϵ |f(f(x))-5|<\epsilon

Note that for any x x chosen in a δ \delta -radius neighborhood of 6, we have that f ( x ) > 2 f(x)>2 , (thereby giving the approaching from the right effect). Substituting this value larger than 2 back into f ( x ) f(x) will return a value that is as close as 5 as you would like (say, ϵ \epsilon close), based on your choice of x x in the interval ( 6 δ , 6 + δ ) (6-\delta, 6+\delta) . Therefore, the limit must be 5.

I know that proof is a bit hand-wavy but I do not have the actual function, just its graph."

Given fact that many people seem confident that answer is DNE, well exactly what is report system for, right?

While you are right about people being wrong from time to time. I know for sure I am correct; Brilliant staffs even verified it. I am sad to see so many disagree with me. That would mean people that viewed discussions will most likely jumping into bandwagon and at the end, they don't learn anything from this problem, possibly like you.

Micah Wood - 5 years, 5 months ago

@Eric Belrose Also, I asked you to sketch f ( f ( x ) ) f(f(x)) . Why didn't you do it?

Micah Wood - 5 years, 5 months ago

Log in to reply

@Micah Wood I did graph it, and I reviewed what you sent. I didn't need to do it, I told you I could easily understand your logic. I just didn't agree.

I think this post helps your cause as well. http://math.stackexchange.com/questions/1589077/problem-of-limit-of-function-of-function.

Thank you for the rough proof. I appreciate the hard work you have put into defending your case.

It is unfortunate you feel sad to be challenged, you should welcome such challenge otherwise why are you on these forums?

I didn't jump on any bandwagon, and I have learned a few things. But you are upset so I understand why you would make such a comment. Remember these forums are to discuss solutions not just to post different solutions that agree with yours. Deep breath, internet on!

Eric Belrose - 5 years, 5 months ago

Log in to reply

@Eric Belrose Here is another example: Same reasoning as you've given. http://math.stackexchange.com/questions/608618/limit-composition

I think I must concede. It still feels wrong to discount the initial function. If the limit doesn't exist at f(L) in the initial function where L is the limit of f(x) as x--> a, it should only be able to have one-sided limits (despite the approach of f(f(x))). A one-sided limit is not the same as having a full limit. Though a one-sided limit exists it isn't the same as being a limit.

Eric Belrose - 5 years, 5 months ago

@Eric Belrose It's not really that I dislike being challenged itself, I am annoyed more of repeatedly being challenged. If they have bother to read some discussions, they wouldn't post disagreement over again and again. (How did my problem get so popular anyway?)

It doesn't feel like discussion here, but more of controversy. You don't see that in other problems...

Since you disagree with my solution, but apparently you agree with answer itself, please write solution, as I am actually bad at explaining something myself.

Micah Wood - 5 years, 5 months ago

That is a very poorly written problem. From the graph you provided one CANNOT say that lim x 6 f ( x ) 2 \lim_{x \to 6}{ f(x)} \neq 2 . I zoomed in the graph before answering the question, and from my point of view f ( x ) = 2 f(x)=2 when x = 6 x=6 . Thus, lim x 6 f ( f ( x ) ) \lim_{x \to 6} {f(f(x))} would not exist since the limits from the right and from the left are different from each other and from f ( 2 ) f(2) .

Rodolfo Marinho - 5 years, 7 months ago

Log in to reply

I was ready to vehemently disagree with this problem when I first read it, until I realized unfortunately that Mr. Woods is actually correct. Looking at the limit from the right and left side of 6: as x-->6+, that is, as the values of x approach 6 from the right, the corresponding values of f(x) are a little more than 2 but getting closer to 2, so the corresponding values of f(f(x)) are a little less than 5 but getting closer to 5. As you look at the limit from the other side, as x--->6-, the corresponding values of f(x) are still a little more than 2, but moving towards 2, so get the same results. As the result, the limit does exist as you get the same value as you approach 6 from the left and right side, namely, you get that the limit is 5. Originally, I assumed that lim x-->6 of f(f(x)) = lim x-->2 of f(x), since f(6)=2, however, I eventually realized that was not true because lim x-->2 implies you are approaching 2 from the right and left side of 2, but this is not the case here, since f(x) is never less than 2 as you approach 6, so you can't use values to the left of 2 as you approach it for final limit evaluation. I hope this explanation clears up any initial misunderstanding that anyone like me originally had. Have a great day.

David Ortiz - 5 years, 3 months ago

OK just report my problem... I find it hilarious that many people dislike this problem, yet nobody reports it...

Meanwhile, lim x 6 f ( f ( x ) ) lim x 2 f ( x ) \displaystyle \lim_{x\to6}f(f(x))\neq\lim_{x\to2}f(x) , because from BOTH right and left side, f ( x ) f(x) approaches 2 from UPWARD. There is no downward. Therefore lim x 6 f ( f ( x ) ) = lim x 2 + f ( x ) \lim_{x\to6}f(f(x)) = \lim_{x\to2^+} f(x) (Note 2 + \mathbf{2^+} ), in other words, f ( x ) f(x) approaches 2 from RIGHT side while x x approaches 6. Hence answer is 5.

I've simulated graph .

Micah Wood - 5 years, 7 months ago

Log in to reply

Apparently many still do not understand the concept of limit, thanks for the problem, great contribution. Greetings from México :)

Angel T. López - 5 years, 5 months ago

How do I report your problem?

Rodolfo Marinho - 5 years, 7 months ago

Log in to reply

@Rodolfo Marinho Click on that \equiv , it's right next to "Calculus Level 2" in top right of problem.

Micah Wood - 5 years, 7 months ago

Log in to reply

@Micah Wood Thank you! Report sent. ;)

Rodolfo Marinho - 5 years, 7 months ago
Calvin Lin Staff
Jun 28, 2016

Approach 1: As x 6 x \rightarrow 6 , we have f ( x ) 2 f(x) \rightarrow 2 . However, the way that f ( x ) f(x) approaches 2 is only on the positive side. Thus, it is more accurate to say that f ( x ) 2 + f(x) \rightarrow 2^+ . Hence,

lim x 6 f ( f ( x ) ) = lim y 2 + f ( y ) = 5 \lim_{x \rightarrow 6} f(f(x) ) = \lim_{y \rightarrow 2^+} f(y) = 5


Approach 2: Let's create a table for x , f ( x ) , f ( f ( x ) ) x, f(x), f(f(x)) (approximate values).

x x 5 5.5 5.9 6 6.1 6.5 7
f ( x ) f(x) 2.2 2.1 2.01 2.01 2.1 2.2
f ( f ( x ) ) f(f(x)) 4.5 4.9 4.99 4.99 4.9 4.5

From the table, we see that the limit is 5.

Why only on positive side in approach 1

Aditya Sahu - 4 years, 11 months ago

can you explain the point of studying f(f(x))? I expect brutal maths explication if you are able to.

Thanks

Vatea Tahiti - 4 years ago
Harry Ray
Jun 27, 2016

For a limit to exist we need the limits from either side to exist, and to be equal.

Since d f d x < 0 \frac{\mathrm{d}f}{\mathrm{d}x} < 0 as x 6 x \to 6^{-} , the left hand limit is: lim x 6 f ( f ( x ) ) = lim x 2 + f ( x ) = 5 \begin{aligned} \lim_{x \to 6^{-}} f(f(x)) &= \lim_{x \to 2^{+}} f(x) \\ &= 5 \end{aligned}

Since d f d x > 0 \frac{\mathrm{d}f}{\mathrm{d}x} > 0 as x 6 + x \to 6^{+} , the right hand limit is: lim x 6 + f ( f ( x ) ) = lim x 2 + f ( x ) = 5 \begin{aligned} \lim_{x \to 6^{+}} f(f(x)) &= \lim_{x \to 2^{+}} f(x) \\ &= 5 \end{aligned}

Therefore, lim x 6 f ( f ( x ) ) = 5 \lim_{x \to 6} f(f(x)) = \boxed{5} .

I had thought of this as evaluating the inner limit first, which would reduce the problem to the limit of f(x) as x approaches 2 which DNE.

However, under the solution's rationale, I find a paradox: Define g(x)=f(f(x)) using f(x) as above except that f(x) = 5 instead of 6. Is g(x) continuous at x=2? Algebraically, the answer is YES because the limit of g(x) as x approaches 2 is 5, and g(2) is also 5. Graphically the answer is NO: Clearly a jump discontinuity. Hmmmm??

Rob Matuschek - 4 years, 4 months ago

Log in to reply

Forget it!! No paradox! Don't read that!!!!!!

Rob Matuschek - 4 years, 4 months ago

Log in to reply

I think you can delete it if you want.

Whitney Clark - 4 years, 3 months ago
Carsten Meyer
Feb 19, 2019

For any who wonder why this is an accurate limit: Remember that limits are defined by small open intervals, later called open sets. The limit does not \color{#D61F06}\text{not}\color{#333333} depend on f ( x 0 ) f(x_0) :

" lim x x 0 f ( x ) = y 0 " " For any (small) ε > 0 : f ( x ) y 0 < ε as long as 0 < x x 0 < δ " \begin{aligned} "\lim_{x\rightarrow x_0}f(x) &= y_0" & \Leftrightarrow && "\text{For any (small) }\varepsilon>0:\quad|f(x) - y_0|<\varepsilon \quad\text{ as long as } \quad\color{#D61F06} 0< \color{#333333}|x-x_0|<\delta" \end{aligned}

Now to the task: We have x 0 = 6 x_0=6 , guess y 0 = 5 y_0=5 and create any small interval I y : = ( 5 ε , 5 + ε ) I_y:=(5-\varepsilon,\:5+\varepsilon) around y 0 = 5 y_0=5 . Let's construct a small interval I x : = ( 6 δ , 6 + δ ) I_x:=(6-\delta,\:6+\delta) using the graphic by the following steps:

  1. Take any (small) ε > 0 \varepsilon>0 and find a function value 5 ε < y 1 < 5 5-\varepsilon < y_1 < 5 .
  2. Find a corresponding x x -value slightly bigger than 2 such that f ( x 1 ) = y 1 f(x_1)=y_1
  3. Make δ > 0 \delta>0 small enough, so that for any 0 < x 6 < δ : f ( x ) 2 < x 1 2 \begin{aligned} 0<|x-6|<\delta:&&&|f(x)-2| < |x_1-2| \end{aligned}

By construction, the following statement holds: " 0 < x 6 < δ f ( f ( x ) ) 5 < ε " lim x 6 f ( f ( x ) ) = 5 \begin{aligned} "0<|x-6|<\delta&&\Rightarrow&&|f(f(x))-5|<\varepsilon"&&\Rightarrow&&\lim_{x\rightarrow 6}f(f(x))=\fbox{5} \end{aligned}

Rem.: You cannot argue with differentiation or continuity here, because f f is not continuous in x = 2 x=2 . However, for a limit to exist we only need to construct these small open intervals, regardless of what properties the function might have.

O limite não existe, pois quando x-->2 ==> lim f(x) é indeterminado. x-->2+ ==> f-->5 x-->2- ==> f-->7

Angelo Hafner - 8 months, 2 weeks ago

Log in to reply

We do not need to consider the case x 2 x\rightarrow 2^- , because lim x 6 f ( x ) = 2 + \displaystyle \lim_{x\rightarrow 6}f(x) = 2^{+} -- regardless from which direction x x tends toward 6 6 !

Carsten Meyer - 2 months, 2 weeks ago

0 pending reports

×

Problem Loading...

Note Loading...

Set Loading...