What is the difference in this limit?

  • Thread starter Thread starter sylent33
  • Start date Start date
  • Tags Tags
    Difference Limit
Click For Summary
The discussion centers on calculating the limit of the function f(x) = (x^2-9)e^{-x} as x approaches both positive and negative infinity. For positive infinity, the limit correctly approaches 0 using L'Hôpital's Rule, but for negative infinity, the limit diverges to infinity, which initially confuses the participants. It is clarified that L'Hôpital's Rule is not applicable when the limit is of the form infinity/0, as this results in infinity or negative infinity depending on the sign of the denominator. The conversation also touches on the abstract nature of infinity in limits, emphasizing that multiplying by constants can affect the outcome. The discussion concludes with a deeper understanding of how limits behave under different conditions, particularly regarding the signs of the terms involved.
sylent33
Messages
39
Reaction score
5
Homework Statement
Calculate the limit of the function for + and - infinity
Relevant Equations
L'Hospital
Hello!

I need to calculate the limit of this function ## f(x) = (x^2-9)*e^{-x}## for + and - ## \infty ## Now for + infinity I did this

$$ \frac{(x^2-9)}{e^x} $$ apply L'Hospital since we have infinity divided by infinity; $$\frac{2x}{e^x} $$ Apply L'Hospital again $$ \frac{2}{e^x} $$ the limit is 0. Now for -##\infty## I did exactly the same,getting 0 as my limit but apperently that is wrong.The answer should be ##\infty## and I don't know how.Intuitively that just does not make sense to me because we are simply approching it from a diffrent side but the same value.Some insight would be great,thanks!
 
  • Like
Likes Delta2
Physics news on Phys.org
The initial limit is not of the form infinity over infinity at minus infinity. What does ##e^x## go to?
 
  • Like
Likes hutchphd
Office_Shredder said:
The initial limit is not of the form infinity over infinity at minus infinity.
@sylent33, this means that L'Hopital's Rule cannot be applied for this limit, assuming that's what you did.
 
  • Like
Likes hutchphd
Uh okay you are right,if we have minus infinity than the ## e^x ## is not going to infinity but rather to 0,so than we have infinity/0. Now I think that is 0,so I don't need to to use the L'Hopitals rule but,this raises another interesting question.So L'Hopitals rule says we need to get the derivative of both the nummerator and determinator.So let's assume I have a situation infinity/0, can I still apply the L'Hopital rule here?
 
Infinity/0 is either positive or negative infinity, depending on the sign of the denominator (or undefined if the denominator keeps flipping signs).L'hopital's rule says you need it to be 0/0 or infinity/infinity in order to take both derivatives. You can't just apply it to arbitrary limits.
 
  • Like
Likes Delta2 and sylent33
Office_Shredder said:
Infinity/0 is either positive or negative infinity, depending on the sign of the denominator (or undefined if the denominator keeps flipping signs).L'hopital's rule says you need it to be 0/0 or infinity/infinity in order to take both derivatives. You can't just apply it to arbitrary limits.
You are right of course infinity/0 is infinity...Making really really silly mistakes.Now since the concept of infinity is very abstract and I know that we cannot consider it a number,but ocassionaly the situation comes in let's say calculating limits of a sequence or a function where we have ## 2 *\infty ## or ##2 * -\infty ## and most of these are usually though in class.But not to long ago I stumbled upon an interesting situation,it was a constant number(negative constant number) times - infinity.Something like this

## -2 * \infty ## and the result was ##-\infty ##.Now I know that we cannot consider infinity to be a number,so I am curious as to how does one approach this,is there a certain method you have to use to evalue this,or is it something you would find on a formula sheet?
 
sylent33 said:
You are right of course infinity/0 is infinity...
Not necessarily. If the denominator is approaching zero through the negative numbers, the limit won't be infinity.
sylent33 said:
Making really really silly mistakes.Now since the concept of infinity is very abstract and I know that we cannot consider it a number,but ocassionaly the situation comes in let's say calculating limits of a sequence or a function where we have ## 2 *\infty ## or ##2 * -\infty ## and most of these are usually though in class.But not to long ago I stumbled upon an interesting situation,it was a constant number(negative constant number) times - infinity.Something like this

## -2 * \infty ## and the result was ##-\infty ##.Now I know that we cannot consider infinity to be a number,so I am curious as to how does one approach this,is there a certain method you have to use to evalue this,or is it something you would find on a formula sheet?
If you have a function whose limit is infinity, what this really means is that the function values get arbitrarily large as the independent variable approaches its limiting value.

So if you're given that ##\lim_{x \to a} f(x) = \infty##, then the following will be true.
##c\lim_{x \to a} f(x) = \infty##, if c is finite and strictly positive.
##c\lim_{x \to a} f(x) = -\infty##, if c is finite and strictly negative.
If c = 0, you can't say what the limit will be (if any) without doing some more work.

Also, if ##\lim_{x \to a} g(x) = \infty##, then the following will be true.
##\lim_{x \to a} \left(f(x) + g(x)\right) = \infty##.
But ##\lim_{x \to a} \left(f(x) - g(x)\right)## is indeterminate, as is ##\lim_{x \to a} \frac{f(x)}{g(x)}##.
 
Mark44 said:
Not necessarily. If the denominator is approaching zero through the negative numbers, the limit won't be infinity.
If you have a function whose limit is infinity, what this really means is that the function values get arbitrarily large as the independent variable approaches its limiting value.

So if you're given that ##\lim_{x \to a} f(x) = \infty##, then the following will be true.
##c\lim_{x \to a} f(x) = \infty##, if c is finite and strictly positive.
##c\lim_{x \to a} f(x) = -\infty##, if c is finite and strictly negative.
If c = 0, you can't say what the limit will be (if any) without doing some more work.

Also, if ##\lim_{x \to a} g(x) = \infty##, then the following will be true.
##\lim_{x \to a} \left(f(x) + g(x)\right) = \infty##.
But ##\lim_{x \to a} \left(f(x) - g(x)\right)## is indeterminate, as is ##\lim_{x \to a} \frac{f(x)}{g(x)}##.
Great answer but I am not sure I get the "if the denominator part is approaching zero through the negative numbers".But than it wouldn't even be infinity/0, but rather infinity/-infinity?
 
Consider ##\lim_{x\to \infty} 1/(e^{-x})## vs ##\lim_{x\to \infty} 1/(-e^{-x})## In both cases it goes to 1/0, but one is infinity and the other is negative infinity. The sign of 0 is not determined.
 
  • Like
Likes sylent33
  • #10
sylent33 said:
Great answer but I am not sure I get the "if the denominator part is approaching zero through the negative numbers".But than it wouldn't even be infinity/0, but rather infinity/-infinity?
No, it wouldn't be of the form infinity/-infinity if the denominator is getting close to zero (but is always negative).
 
Last edited:
  • #11
I think I get it now,shredder gave an great example