The Derivative of Bessel Function of the Second Kind

Click For Summary

Discussion Overview

The discussion revolves around the derivative of the Bessel function of the second kind, specifically the expression \(\frac{d}{dx}K_v(f(x))\). Participants explore the application of the chain rule and related identities in the context of Bessel functions, addressing both theoretical and practical aspects of differentiation with more complex arguments.

Discussion Character

  • Technical explanation
  • Mathematical reasoning
  • Debate/contested

Main Points Raised

  • One participant asks for the derivative of \(K_v(f(x))\) without providing a specific method.
  • Another participant references an identity involving \(C_{\nu}(y)\) and suggests using the chain rule after substituting \(y = f(x)\).
  • A participant expresses confusion regarding the application of known identities when dealing with more complex arguments, such as \(z = \sqrt{x^2+x}\), and proposes a method to transition from simpler to more complex cases.
  • There is a reiteration of the identity \(z\frac{d}{dz}K_v(z)+vK_v(z)=-zK_{v-1}(z)\) and an attempt to adapt it for \(f(z)\) with a proposed expression, seeking validation from others.
  • Some participants affirm the correctness of the proposed method and expression, while others continue to seek clarity on the derivative's evaluation.

Areas of Agreement / Disagreement

Participants express varying levels of understanding and confidence regarding the derivative of the Bessel function with complex arguments. There is no clear consensus on the final expression or method, and some participants remain uncertain about the correctness of their approaches.

Contextual Notes

Participants reference identities and methods that may depend on specific definitions or assumptions about the functions involved. The discussion includes unresolved mathematical steps and varying interpretations of how to apply the chain rule in this context.

EngWiPy
Messages
1,361
Reaction score
61
Hello,

What is [tex]\frac{d}{dx}K_v\left(f(x)\right)=?[/tex]

Thanks in advance
 
Physics news on Phys.org
For [itex]\nu[/itex] not necessarily an integer, [itex]C_{\nu}(y) = e^{\nu \pi i}K_{\nu}(y)[/itex] satisfies the identity

[tex]2\frac{dC_\nu}{dy} = C_{\nu-1}(y) + C_{\nu + 1}(y)[/tex]

Then let [itex]y = f(x)[/itex] and use the chain rule.

Identities like this can usually be found on wikipedia, and for something as studied and used as frequently as the Bessel Functions, are generally correct.

http://en.wikipedia.org/wiki/Bessel_function

For a more 'official' reference, see something like http://www.math.sfu.ca/~cbm/aands/page_437.htm (scans of a reference book).
 
Mute said:
For [itex]\nu[/itex] not necessarily an integer, [itex]C_{\nu}(y) = e^{\nu \pi i}K_{\nu}(y)[/itex] satisfies the identity

[tex]2\frac{dC_\nu}{dy} = C_{\nu-1}(y) + C_{\nu + 1}(y)[/tex]

Then let [itex]y = f(x)[/itex] and use the chain rule.

Identities like this can usually be found on wikipedia, and for something as studied and used as frequently as the Bessel Functions, are generally correct.

http://en.wikipedia.org/wiki/Bessel_function

For a more 'official' reference, see something like http://www.math.sfu.ca/~cbm/aands/page_437.htm (scans of a reference book).

I know that
[tex]z\frac{d}{dz}K_v(z)+vK_v(z)=-zK_{v-1}(z)[/tex]
but I was confused when we have more complicated arguments such as
[tex]z=\sqrt{x^2+x}[/tex].
But after your posting, I have now a simple method to move from simple to more complicated arguments. So, I can say the following:

[tex]\frac{f(z)}{f'(z)}\frac{d}{dz}K_v(f(z))+vK_v(f(z))=-f(z)K_{v-1}(f(z))[/tex]

Am I right?

Best regards
 
S_David said:
Hello,

What is [tex]\frac{d}{dx}K_v\left(f(x)\right)=?[/tex]

Thanks in advance

Do you get the answer?
 
skyspeed said:
Do you get the answer?

let [tex]y=f(x)[/tex] and then use the chain rule.
 
S_David said:
I know that
[tex]z\frac{d}{dz}K_v(z)+vK_v(z)=-zK_{v-1}(z)[/tex]
but I was confused when we have more complicated arguments such as
[tex]z=\sqrt{x^2+x}[/tex].
But after your posting, I have now a simple method to move from simple to more complicated arguments. So, I can say the following:

[tex]\frac{f(z)}{f'(z)}\frac{d}{dz}K_v(f(z))+vK_v(f(z))=-f(z)K_{v-1}(f(z))[/tex]

Am I right?

Best regards

i think this is right
 
S_David said:
let [tex]y=f(x)[/tex] and then use the chain rule.

thanks a lot
 

Similar threads

  • · Replies 3 ·
Replies
3
Views
3K
  • · Replies 3 ·
Replies
3
Views
2K
  • · Replies 1 ·
Replies
1
Views
1K
  • · Replies 4 ·
Replies
4
Views
2K
  • · Replies 2 ·
Replies
2
Views
2K
  • · Replies 10 ·
Replies
10
Views
3K
  • · Replies 6 ·
Replies
6
Views
3K
  • · Replies 3 ·
Replies
3
Views
2K
  • · Replies 4 ·
Replies
4
Views
4K
  • · Replies 4 ·
Replies
4
Views
3K