Legendre-Gauss Quadrature formula: Difference between revisions

From Citizendium
Jump to navigation Jump to search
imported>Dmitrii Kouznetsov
(Did not finish yet.)
imported>Dmitrii Kouznetsov
Line 7: Line 7:
Legendre-Gauss quadratude formula is special case of [[Gaussian quadratures]] of more general kind, which allow efficient approximation of a function with known asumptiotic behavior at the edges of the interval of integration.
Legendre-Gauss quadratude formula is special case of [[Gaussian quadratures]] of more general kind, which allow efficient approximation of a function with known asumptiotic behavior at the edges of the interval of integration.
==Nodes and weights==
==Nodes and weights==
Nodes <math>x_i</math> in equation (1) are zeros of the [[Polunomial of Lehendre]]  <math>P_N</math>:
Nodes <math>x_i</math> in equation (1) are zeros of the [[Polynomial of Legendre]]  <math>P_N</math>:
: (2) <math> P_N(x_i)=0</math>  
: (2) <math> P_N(x_i)=0</math>  
: (3) <math> -1<x_1<x_2< ... <x_N <1</math>
: (3) <math> -1<x_1<x_2< ... <x_N <1</math>
Line 31: Line 31:
|publisher=[[Do not forget to type data of the ref. here!]]
|publisher=[[Do not forget to type data of the ref. here!]]
}}</ref>
}}</ref>
==Precision of the approximation==
==Precision of the approximation==



Revision as of 09:40, 27 May 2008

Template:Copyedit

Legendre-Gauss Quadratude formiula is the approximation of the integral

(1)

with special choice of nodes and weights , characterised in that, if the finction is polynomial of order smallet than , then the exact equality takes place in equation (1).

Legendre-Gauss quadratude formula is special case of Gaussian quadratures of more general kind, which allow efficient approximation of a function with known asumptiotic behavior at the edges of the interval of integration.

Nodes and weights

Nodes in equation (1) are zeros of the Polynomial of Legendre :

(2)
(3)

Weight in equaiton (1) can be expressed with

(4)

There is no straightforward espression for the nodes ; they can be approximated with many decimal digits through only few iterations, solving numerically equation (2) with initial approach

(5)

These formulas are described in the books [1] [2]

Precision of the approximation

Example

Fig.1. Example of estimate of precision: Logarithm of residual versus number of terms in the right hand side of equation (1) for various integrands .

Extension to other interval

is straightforward. Should I copypast the obvious formulas here?

References

  1. (1980) Handbook on special functions. 
  2. Numerical Resipes in C. Do not forget to type data of the ref. here!.