Taylor Series Expansion of f(x) at 0

Click For Summary

Homework Help Overview

The discussion revolves around finding the Taylor series expansion of a function f(x) at 0, specifically focusing on the computation of derivatives and the series representation of the function. Participants are exploring the complexities involved in deriving higher-order derivatives and the potential for alternative methods to simplify the process.

Discussion Character

  • Exploratory, Conceptual clarification, Mathematical reasoning

Approaches and Questions Raised

  • Participants are attempting to compute the first few derivatives of f(x) and are questioning the difficulty of obtaining higher-order derivatives. Some suggest using series expansions for sine and hyperbolic cosine functions as an alternative approach. Others discuss the implications of the function being odd and how that affects the derivatives.

Discussion Status

The discussion is active with various participants offering different methods and insights. Some have provided suggestions for simplifying the process, while others are clarifying misunderstandings about the original post and its edits. There is no explicit consensus on the best approach, but multiple lines of reasoning are being explored.

Contextual Notes

Participants note the importance of the function's properties, such as being odd, which influences the derivatives. There is also mention of the original post being edited, which may have affected the clarity of the discussion.

esrever10
Messages
1
Reaction score
0
Homework Statement
Expand the function ##f(x)=\sin x / (\cosh x + 2)## in a Taylor series around the origin going up to ##x^3##
Relevant Equations
##f(x)=f(a)+f^{(1)}(a)(x-a)+\frac{1}{2!}f(2)(a)(x-a)^2+...##
First I got ##f(0)=0##,

Then I got ##f'(x)(0)=\frac{\cos x(2+\cosh x)-\sin x\sinh x}{(2+\cosh x)^2}=1/3##

But when I tried to got ##f''(x)## and ##f'''(x)##, I felt that's terrible, If there's some easy way to get the anwser?
 
Last edited:
Physics news on Phys.org
Please show your work.
 
  • Like
Likes   Reactions: esrever10
Orodruin said:
Please show your work.
I don't get this, they showed us how they computed the first term of the series.

Esrever, a neat trick is you can write out the numerator and denominator as series, then apply ##\frac{1}{1-y}=1+y+y^2...## to the denominator where ##y## needs to be all the stuff that goes to zero. Then a bit of algebra on what's left gets you the solution.

This is not that much easier, but probably feels more fulfilling than computing tedious derivatives.
 
  • Like
Likes   Reactions: esrever10, FactChecker and topsquark
Office_Shredder said:
I don't get this, they showed us how they computed the first term of the series.
If you look at the last edit time of the OP you will see that it is later than when I posted. The original OP did not include this.
 
  • Like
Likes   Reactions: malawi_glenn
Orodruin said:
If you look at the last edit time of the OP you will see that it is later than when I posted. The original OP did not include this.

Sorry, my mistake for missing that.
 
  • Like
Likes   Reactions: malawi_glenn
If you find the series for sin(x) and cosh(x) out to enough terms, then you can do regular multiplication and long division of polynomials.
 
  • Like
Likes   Reactions: esrever10 and topsquark
You already have f(0) and f'(0). You (should) know that f''(0) = 0 because f is odd, so all you need is f'''(0). What you don't need is the general expression for f'''(x) in terms of x; an expression for f'''(0) in terms of f(0), f'(0) and f''(0) will suffice.

Start with <br /> (2 + \cosh x)f(x) = \sin x and differentiate both sides three times using the product rule: <br /> \sum_{n=0}^3 \binom{3}{n} f^{(n)}(x) \frac{d^{3-n}}{dx^{3-n}}(2 + \cosh x) = -\cos x. Now substitute x = 0 and solve for f&#039;&#039;&#039;(0).
 
  • Like
Likes   Reactions: esrever10
I'd probably go with the method of post #3. I do find it simpler than brute forcing by computing the derivatives.
 

Similar threads

Replies
2
Views
1K
Replies
4
Views
2K
  • · Replies 18 ·
Replies
18
Views
3K
  • · Replies 27 ·
Replies
27
Views
3K
Replies
2
Views
2K
  • · Replies 2 ·
Replies
2
Views
2K
Replies
6
Views
3K
  • · Replies 1 ·
Replies
1
Views
2K
Replies
5
Views
2K
  • · Replies 5 ·
Replies
5
Views
3K