| 
November 10th, 2008, 04:15 AM
| | Junior Member | | Join Date: Nov 2008
Posts: 8
Country: Thanks: 0
Thanked 0 Times in 0 Posts
| | Big Oh in arithmetic expression Hello to everyone,
Could somebody please tell me how do we solve exercises like the one I describe below: = e + O(1/n)
Thanks a lot,
Tasos | 
November 10th, 2008, 04:29 AM
| | Member | | Join Date: Oct 2008 Location: Seattle, Washington
Posts: 62
Country: Thanks: 0
Thanked 19 Times in 19 Posts
| | Hi Tasos: I see the equation, but I do not know what the instructions are for this exercise. Are exp(1) and e both the same constant in this equation?
What are we to do with this equation?
~ Mark | 
November 10th, 2008, 04:34 AM
| | Junior Member | | Join Date: Nov 2008
Posts: 8
Country: Thanks: 0
Thanked 0 Times in 0 Posts
| | First of all, I want to thank you mmmbot for answering.
Sorry if I didn't write it well. By writing exp I meant e to the power.
So I mean: [e to the power of (1 + (O(1/n))^2)] = e + O(1/n)
P.S.1: Could you tell me how to write it in the post so as it will appear correctly?
P.S.2: What you say is right: exp(1) and e are the same thing. That is what I mean. | 
November 10th, 2008, 04:36 AM
| | Member | | Join Date: Oct 2008 Location: Seattle, Washington
Posts: 62
Country: Thanks: 0
Thanked 19 Times in 19 Posts
| | I read it as follows. e^[1 + (O/n)^2] = e + O/n | 
November 10th, 2008, 04:44 AM
| | Junior Member | | Join Date: Nov 2008
Posts: 8
Country: Thanks: 0
Thanked 0 Times in 0 Posts
| | Quote:
Originally Posted by mmm4444bot I read it as follows. e^[1 + (O/n)^2] = e + O/n |
Sorry, then. I didn't write O/n, but O(1/n).
Is it the same thing? | 
November 10th, 2008, 04:53 AM
| | Member | | Join Date: Oct 2008 Location: Seattle, Washington
Posts: 62
Country: Thanks: 0
Thanked 19 Times in 19 Posts
| | I'm checking to see whether or not O(1/n) is function notation ... be right back. A big "Oh, please excuse me". 
I thought O and n were constants. (I just saw the big Oh function for the first time at Wikipedia.)
No, you are correct. O(1/n) is good.
I cannot help you with this exercise.
Now, I read it as follows.
e^[1 + O(1/n)^2] = e + O(1/n) Again, sorry for goofing up your thread.
~ Mark | 
November 10th, 2008, 05:04 AM
| | Junior Member | | Join Date: Nov 2008
Posts: 8
Country: Thanks: 0
Thanked 0 Times in 0 Posts
| | Quote:
Originally Posted by mmm4444bot I'm checking to see whether or not O(1/n) is function notation ... be right back. A big "Oh, please excuse me". 
I thought O and n were constants. (I just saw the big Oh function for the first time at Wikipedia.)
No, you are correct. O(1/n) is good.
I cannot help you with this exercise.
Now, I read it as follows.
e^[1 + O(1/n)^2] = e + O(1/n) Again, sorry for goofing up your thread.
~ Mark | No problem, Mark. Don't worry. | 
November 10th, 2008, 12:36 PM
| | Super Member | | Join Date: Aug 2008 Location: Lyon, France
Posts: 612
Country: Thanks: 34
Thanked 391 Times in 328 Posts
| | Quote:
Originally Posted by tasos Hello to everyone,
Could somebody please tell me how do we solve exercises like the one I describe below: = e + O(1/n)
Thanks a lot,
Tasos | Are you sure you put the parentheses at the right place? Because this is correct, but not optimal, as you shall see:
I'll apply usual properties of the "big O" notation. If you don't know them, just ask, the proofs are very short.
First you have  , and it is more usual to write it this way.
Then  .
I used the following expansion of the exponential at 0:  when  tends to 0, composed with the sequence  which tends to 0.
Now,  , hence the previous big O can be replaced by  , and this is what you need.
Let us now suppose that the question was  , which seems more plausible to me.
Then we would have:  (as before, the second big O can be included in the first one), and by the same computation as above we conclude  .
ps: about the way to write symbols, there's a section in the forum related to "LaTeX", and this is where you should look for help about this. | 
November 10th, 2008, 03:44 PM
| | Junior Member | | Join Date: Nov 2008
Posts: 8
Country: Thanks: 0
Thanked 0 Times in 0 Posts
| | Thank you very much Laurent for your answer. It is very thorough.
However, I have a little question.  How do you go from the left part to the right one in the following expression: Quote:
Originally Posted by Laurent |
Thanks | 
November 11th, 2008, 03:57 AM
| | Super Member | | Join Date: Aug 2008 Location: Lyon, France
Posts: 612
Country: Thanks: 34
Thanked 391 Times in 328 Posts
| | Quote:
Originally Posted by tasos Thank you very much Laurent for your answer. It is very thorough.
However, I have a little question.  How do you go from the left part to the right one in the following expression:
Thanks | This is because of this result: Quote:
Originally Posted by Laurent I used the following expansion of the exponential at 0:  when  tends to 0, composed with the sequence  which tends to 0. | In fact, as soon as a function  is differentiable at  , we have  as  tends to 0 (because  so that  as  ).
And if  as  , then  , so that we can compose:  . This can be written  as  tends to 0.
Now, in this expansion, you can replace  by any sequence which converges to 0. For instance,  . If  , you get what I wrote and used.
(I'm thinking of something that may have been confusing: when I wrote  , it meant  , not exponential of the parenthesis) | 
November 16th, 2008, 09:14 AM
| | Junior Member | | Join Date: Nov 2008
Posts: 8
Country: Thanks: 0
Thanked 0 Times in 0 Posts
| | Laurent, I want to thank you very much for your answers. They were very comprehensible. | | Thread Tools | | | | Display Modes | Linear Mode |
Posting Rules
| You may not post new threads You may not post replies You may not post attachments You may not edit your posts HTML code is Off | | | All times are GMT -7. The time now is 06:31 PM. | | |