What is the next step for part iii in Taylor Series Extrapolation?

Click For Summary

Discussion Overview

The discussion revolves around the steps needed to solve part iii of a problem related to Taylor Series Extrapolation, focusing on achieving second-order accuracy in approximating a function. Participants explore the mathematical conditions required for the coefficients in the extrapolation formula.

Discussion Character

  • Technical explanation
  • Mathematical reasoning
  • Debate/contested

Main Points Raised

  • One participant requests guidance on how to proceed with part iii of the problem.
  • Another participant outlines the conditions for second-order accuracy, presenting equations that involve derivatives evaluated at a specific point.
  • A participant seeks clarification on the notation used and questions whether a third equation would be necessary for third-order accuracy.
  • Clarification is provided regarding the notation and the implications of achieving third-order accuracy.
  • Participants discuss the values of coefficients a and b, with one participant claiming to have found a as 1 and expressing uncertainty about b.
  • Another participant reminds that since h=0, terms involving h should not appear in the final expressions.
  • A different approach is suggested, involving Taylor expansions, to derive the necessary conditions for a and b.
  • One participant acknowledges a mistake in including h^2 and proposes new values for a and b based on their calculations.
  • A later reply affirms the proposed values for a and b as correct.

Areas of Agreement / Disagreement

Participants express differing views on the values of coefficients a and b, with some proposing different values and approaches. The discussion remains unresolved regarding the exact values of a and b, as participants are still working through the problem.

Contextual Notes

Participants have not reached a consensus on the final values of a and b, and there are unresolved mathematical steps in the derivation process.

ajd-brown
Messages
30
Reaction score
0
Hi could anyone give me pointer as to where to go with part iii please?
 

Attachments

  • question.jpg
    question.jpg
    22.9 KB · Views: 552
  • Solution so far.jpg
    Solution so far.jpg
    18.7 KB · Views: 552
Physics news on Phys.org
second order accurate means
$$\begin{cases}
A|_{h=0}=B|_{h=0}
\\
\left. \dfrac{dA}{dh}\right|_{h=0}=\left. \dfrac{dB}{dh}\right|_{h=0}
\end{cases}$$
so require
$$\begin{cases}
\mathrm{f}(x+2h)|_{h=0}=a \, \mathrm{f}(x+h)+b \, \mathrm{f}(x)|_{h=0}
\\
\left. \dfrac{d}{dh}\mathrm{f}(x+2h)\right|_{h=0}=\left. \dfrac{d}{dh}(a \, \mathrm{f}(x+h)+b \, \mathrm{f}(x))\right|_{h=0}
\end{cases}$$
then solve for a and b
 
  • Like
Likes   Reactions: 1 person
Hi, I don't really understand the notation you have used there with the lines, but does that basically mean that the equation holds true when differentiated once and in its original form?

so just to check my understanding, if it was supposed to be 3rd order accurate, there would be a third equation which would be the second differential of the original and this would have to be satisfied as well by the values of a and b?
 
The line means let the variable take a certain value
in this case let h=0 after taking the derivative

yes third order accurate would be
\begin{cases}
A|_{h=0}=B|_{h=0}
\\
\left. \dfrac{dA}{dh}\right|_{h=0}=\left. \dfrac{dB}{dh}\right|_{h=0}
\\
\left. \dfrac{d^2A}{dh^2}\right|_{h=0}=\left. \dfrac{d^2B}{dh^2}\right|_{h=0}
\end{cases}
and so on
 
  • Like
Likes   Reactions: 1 person
did you find a and b?
 
Ok that's great thank you! I am going to attempt it now that I understand it! I'll place my answer on here afterwards, I gather you have found them then?
 
^I could not resist
 
I have found a as 1 is that correct? and using that value I get b as

b=((((hdf(x)/dx + 3/2*h^2*d2f(x)/dx))))/f(x)
 
Last edited:
Remember h=0 so it should not appear. Since we are second order we should not have h^2.

Here is another equivalent approach

$$\text{if}
\\
\mathrm{f}(x+2h)=\mathrm{f}(x)+2h\, \dfrac{df}{dx}(x)+\mathrm{O}(x^2)
\\
\mathrm{f}(x+h)=\mathrm{f}(x)+h\, \dfrac{df}{dx}(x)+\mathrm{O}(x^2)
\\
\mathrm{f}(x+2h)=a\, \mathrm{f}(x+h)+b\, \mathrm{f}(x)
\\
\text{then}
\\
\mathrm{f}(x)+2h\, \dfrac{df}{dx}(x)+\mathrm{O}(x^2)=a\left[\mathrm{f}(x)+h\, \dfrac{df}{dx}(x)\right]+b\, \mathrm{f}(x)+\mathrm{O}(x^2)
\\
\mathrm{f}(x)+2h\, \dfrac{df}{dx}(x)=a\left[\mathrm{f}(x)+h\, \dfrac{df}{dx}(x)\right]+b\, \mathrm{f}(x)$$
What do a and b need to be?
 
  • #10
that makes so much more sense! as you say I included the h^2 accidentally so that must be where i went wrong, I'll try and solve it the original way again and see if I get the same answer!

From what I can see a needs to be 2 and b has to be -1
 
  • #11
^Good job
 

Similar threads

  • · Replies 4 ·
Replies
4
Views
3K
  • · Replies 17 ·
Replies
17
Views
4K
  • · Replies 5 ·
Replies
5
Views
17K
  • · Replies 3 ·
Replies
3
Views
3K
  • · Replies 11 ·
Replies
11
Views
2K
Replies
1
Views
2K
  • · Replies 9 ·
Replies
9
Views
3K
  • · Replies 2 ·
Replies
2
Views
2K
  • · Replies 8 ·
Replies
8
Views
3K
  • · Replies 5 ·
Replies
5
Views
2K