satlas.stats.fitting.chisquare_model¶

satlas.stats.fitting.chisquare_model(params, f, x, y, yerr, xerr=None, func=None)[source]¶

Model function for chisquare fitting routines as established in this module.

Other Parameters:
Parameters:	params (lmfit.Parameters) – Instance of lmfit.Parameters object, to be assigned to the model object. f (`BaseModel`) – Callable instance with the correct methods for the fitmethods. x (array_like) – Experimental data for the x-axis. y (array_like) – Experimental data for the y-axis. yerr (array_like) – Experimental errorbars on the y-axis.
	xerr (array_like, optional) – Given an array with the same size as x, the error is taken into account by using the method of estimated variance. Defaults to None. func (function, optional) – Given a function, the errorbars on the y-axis is calculated from the fitvalue using this function. Defaults to None.
Returns:	Array containing the residuals for the given parameters, divided by the uncertainty.
Return type:	NumPy array

Note

If a custom function is to be used for the calculation of the residual, this function should be overwritten.

The method of estimated variance calculates the chisquare in the following way:

$\sqrt{\chi^2} = \frac{y-f(x)}{\sqrt{\sigma_x^2+f'(x)^2\sigma_x^2}}$