Chandra X-Ray Observatory
	(CXC)
Skip to the navigation links
Last modified: December 2015

URL: http://cxc.harvard.edu/sherpa/ahelp/plot_pvalue.html
Jump to: Description · Examples · Bugs · See Also


AHELP for CIAO 4.9 Sherpa v1

plot_pvalue

Context: plotting

Synopsis

Plot a histogram of likelihood ratio test results.

Syntax

plot_pvalue(null_model, alt_model [, conv_model=None, id=1,
otherids=(), num=500, bins=25, numcores=None, replot=False,
overplot=False, clearwindow=True])

Description

There are several functions available in Sherpa for performing a likelihood ratio test to compare a fit to data done with a simple, null model versus a more complex, alternative model. The plot_pvalue() function plots a histogram of likelihood ratios comparing fits done with a specified null model to fits done with the alternative model, using data simulated with Poisson noise. It computes the likelihood ratio and the p-value (value used to reject or accept the null model) using the observed data.

The Sherpa fit statistic must be set a maximum-likelihood one (e.g. Cash, CStat, or WStat) for the likelihood ratio test, using the set_stat() command.

Arguments:

  • null_model - model representing the null hypothesis
  • alt_model - alternative model to compare to null
  • conv_model - convolution model (PSF or RSP) to include for fitting; default = None
  • id - Sherpa data id; default = default data id
  • otherids - List of other Sherpa data ids; default = ()
  • num - Number of iterations to run; default = 500
  • bins - Number of bins for the histogram; default = 25
  • numcores - Number of cpus to use during simulation; default = number of detected cpus
  • replot - Send cached data arrays to visualizer; default = False
  • overplot - Plot data without clearing previous plot; default = False

For the likelihood ratio test to be valid, the following conditions must be true:

  • the null model is nested within the more complex, alternative model;
  • the extra parameters of the alternative model have Gaussian (normal) distributions that are not truncated by parameter space boundaries.

Example 1

sherpa> plot_pvalue(powlaw1d.p1, bpl1d.bp1)

Generate a histogram of likelihood ratios comparing fits to simulated data done with a simple, null power-law model to those done with a more-complex, broken power-law model. Do not include a response or PSF convolution model in the fit; use the default number of simulations in the test (500); and use the default number of bins in the resulting histogram of ratios (25).

Example 2

sherpa> rsp1 = get_response()
sherpa> plot_pvalue(p1, p1+g1, conv_model=rsp1, num=600)

Check to see if the addition of a gaussian line to the null power-law model is significant, by applying the likelihood ratio test to compare a power-law (p1) with a power-law plus a line (p1+g1). Plot the distribution of the ratio of (likelihood with line)/(likelihood with no line) for the ensemble of simulations. Include the ARF*RMF instrument response associated with the observed data in the fit, and increase the number of simulations used from default 500 to 600.

Bugs

See the bugs pages on the Sherpa website for an up-to-date listing of known bugs.

See Also

contrib
get_data_prof, get_data_prof_prefs, get_delchi_prof, get_delchi_prof_prefs, get_fit_prof, get_model_prof, get_model_prof_prefs, get_resid_prof, get_resid_prof_prefs, get_source_prof, get_source_prof_prefs, plot_chart_spectrum, prof_data, prof_delchi, prof_fit, prof_fit_delchi, prof_fit_resid, prof_model, prof_resid, prof_source
data
get_arf_plot, get_bkg_plot
info
list_model_ids, show_bkg_model, show_bkg_source
modeling
add_model, add_user_pars, clean, create_model_component, delete_bkg_model, delete_model, delete_model_component, get_model, get_model_autoassign_func, get_model_component, get_model_component_image, get_model_component_plot, get_model_plot, get_num_par, get_order_plot, get_par, get_pileup_model, get_source, get_source_component_image, get_source_component_plot, image_model, image_model_component, image_source, image_source_component, integrate, integrate1d, link, load_table_model, load_template_model, load_user_model, normal_sample, reset, save_model, save_source, set_bkg_model, set_full_model, set_model_autoassign_func, set_pileup_model, set_source, set_xsabund, set_xscosmo, set_xsxsect, set_xsxset, t_sample, uniform_sample
plotting
get_energy_flux_hist, get_lrt_plot, get_lrt_results, get_photon_flux_hist, get_pvalue_plot, get_pvalue_results, get_split_plot, plot, plot_arf, plot_bkg, plot_cdf, plot_chisqr, plot_data, plot_delchi, plot_energy_flux, plot_fit, plot_model, plot_model_component, plot_order, plot_pdf, plot_photon_flux, plot_ratio, plot_resid, plot_scatter, plot_source, plot_source_component, plot_trace, set_xlinear, set_xlog, set_ylinear, set_ylog
psfs
delete_psf, load_conv, plot_kernel
saving
save_delchi, save_resid
statistics
get_chisqr_plot, get_delchi_plot
utilities
calc_chisqr, calc_energy_flux, calc_model_sum, calc_photon_flux, calc_source_sum, calc_stat, eqwidth
visualization
contour_model, contour_ratio, contour_resid, get_ratio, get_resid

Last modified: December 2015
Smithsonian Institute Smithsonian Institute

The Chandra X-Ray Center (CXC) is operated for NASA by the Smithsonian Astrophysical Observatory. 60 Garden Street, Cambridge, MA 02138 USA.   Email:   cxchelp@head.cfa.harvard.edu Smithsonian Institution, Copyright © 1998-2017. All rights reserved.