t_sample

sherpa.astro.ui.t_sample(num=1, dof=None, id=None, otherids=(), numcores=None)

Sample the fit statistic by taking the parameter values from a Student’s t-distribution.

For each iteration (sample), change the thawed parameters by drawing values from a Student’s t-distribution, and calculate the fit statistic.

Parameters:
  • num (int, optional) – The number of samples to use (default is 1).
  • dof (optional) – The number of degrees of freedom to use (the default is to use the number from the current fit).
  • id (int or str, optional) – The data set containing the model expression. If not given then the default identifier is used, as returned by get_default_id.
  • otherids (sequence of int or str, optional) – For when multiple source expressions are being used.
  • numcores (optional) – The number of CPU cores to use. The default is to use all the cores on the machine.
Returns:

A NumPy array table with the first column representing the statistic and later columns the parameters used.
Return type:

samples

See also

fit()
Fit a model to one or more data sets.
normal_sample()
Sample from the normal distribution.
set_model()
Set the source model expression for a data set.
set_stat()
Set the statistical method.
uniform_sample()
Sample from a uniform distribution.

Examples

The model fit to the default data set has three free parameters. The median value of the statistic calculated by t_sample is returned:

>>> ans = t_sample(num=10000)
>>> ans.shape
(1000, 4)
>>> np.median(ans[:,0])
119.9764357725326