Peakpvoigt1: Difference between revisions

From Eigenvector Research Documentation Wiki
Jump to navigation Jump to search
imported>Neal
imported>Scott
 
Line 30: Line 30:


===Algorithm===
===Algorithm===
TEST <math>\sum_{n=0}^\infty \frac{x^n}{n!}</math>
wikipedia <math>f\left( {{a}_{i}},\mathbf{x} \right)={{x}_{1}}\left[ {{x}_{4}}{{\operatorname{e}}^{\frac{-4\ln \left( 2 \right){{\left( {{a}_{i}}-{{x}_{2}} \right)}^{2}}}{x_{3}^{2}}}}+\left( 1-{{x}_{4}} \right)\left[ \frac{x_{3}^{2}}{{{\left( {{a}_{i}}-{{x}_{2}} \right)}^{2}}+x_{3}^{2}} \right] \right]</math>
texvc <math>\[f\left( {{a}_{i}},\mathbf{x} \right)={{x}_{1}}\left[ {{x}_{4}}{{\operatorname{e}}^{\frac{-4\ln \left( 2 \right){{\left( {{a}_{i}}-{{x}_{2}} \right)}^{2}}}{x_{3}^{2}}}}+\left( 1-{{x}_{4}} \right)\left[ \frac{x_{3}^{2}}{{{\left( {{a}_{i}}-{{x}_{2}} \right)}^{2}}+x_{3}^{2}} \right] \right]\]</math>


===Example===
===Example===

Latest revision as of 15:20, 24 October 2013

Purpose

Outputs a pseudo-Voigt function, Jacobian, and Hessian for a given set of input parameters and axis.

Synopsis

[y,y1,y2] = peakpvoigt1(x,ax)

Description

Given a 4-element vector of parameters (x) and a vector of independent variables e.g. a wavelength or frequency axis (ax). PEAKPVOIGT1 outputs a pseudo-voigt peak (y). If more than one output is requested, it also outputs the Jacobian (y1) and Hessian (y2). Derivatives are with respect to the parameters and are evaluated at (x). This function is called by PEAKFUNCTION.

Inputs

  • x = 4 element vector with parameters
  • x(1): coefficient,
  • x(2): mean,
  • x(3): spread, and
  • x(4): fraction Gaussian.
  • ax = 1 by N vector of independent variables e.g,. a wavelength or frequency axis.

Outputs

  • y = 1 by N vector with the pseudo-voigt function evaluated at (x) and given by
z = p-x(2)
y = x(1)*( x(4)*exp(-4*ln(2)*z.^2/x(3)^2) + (1-x(4))*x(3)^2./(x(3)^2+z.^2) );
  • y1 = dy/dxi, 4 by N matrix of the Jacobian of evaluated at (x).
  • y2 = d2y/dxi^2, 4 by 4 by N matrix of the Hessian of evaluated at (x).

Algorithm

Example

Make a single known peak

  ax = 0:0.1:100;
  y  = peakpvoigt1([2 51 8 0.5],ax);
  figure, plot(ax,y)

See Also

peakfunction, peakgaussian, peaklorentzian, peakpvoigt2, peakstruct