WStatCountsStatistic¶

class gammapy.stats.WStatCountsStatistic(n_on, n_off, alpha, mu_sig=None)[source]

Bases: gammapy.stats.counts_statistic.CountsStatistic

Class to compute statistics (significance, asymmetric errors , ul) for Poisson distributed variable with unknown background.

Parameters
n_onint

Measured counts in signal (ON) region

n_offint

Measured counts in background only (OFF) region

alphafloat

Acceptance ratio of ON and OFF measurements

mu_sigfloat

Expected counts in signal region

Attributes Summary

 TS_max Stat value for best fit hypothesis, i.e. TS_null Stat value for null hypothesis, i.e. background counts_off_normalised delta_ts Return TS difference of measured excess versus no excess. error Approximate error from the covariance matrix. excess p_value Return p_value of measured excess. significance Return statistical significance of measured excess.

Methods Summary

 compute_errn([n_sigma]) Compute downward excess uncertainties. compute_errp([n_sigma]) Compute upward excess uncertainties. compute_upper_limit([n_sigma]) Compute upper limit on the signal. excess_matching_significance(significance) Compute excess matching a given significance.

Attributes Documentation

TS_max

Stat value for best fit hypothesis, i.e. expected signal mu = n_on - alpha * n_off - mu_sig

TS_null

Stat value for null hypothesis, i.e. mu_sig expected signal counts

background
counts_off_normalised
delta_ts

Return TS difference of measured excess versus no excess.

error

Approximate error from the covariance matrix.

excess
p_value

Return p_value of measured excess.

significance

Return statistical significance of measured excess.

Methods Documentation

compute_errn(n_sigma=1.0)

Compute downward excess uncertainties.

Searches the signal value for which the test statistics is n_sigma**2 away from the maximum.

Parameters
n_sigmafloat

Confidence level of the uncertainty expressed in number of sigma. Default is 1.

compute_errp(n_sigma=1)

Compute upward excess uncertainties.

Searches the signal value for which the test statistics is n_sigma**2 away from the maximum.

Parameters
n_sigmafloat

Confidence level of the uncertainty expressed in number of sigma. Default is 1.

compute_upper_limit(n_sigma=3)

Compute upper limit on the signal.

Searches the signal value for which the test statistics is n_sigma**2 away from the maximum or from 0 if the measured excess is negative.

Parameters
n_sigmafloat

Confidence level of the upper limit expressed in number of sigma. Default is 3.

excess_matching_significance(significance)

Compute excess matching a given significance.

This function is the inverse of significance.

Parameters
significancefloat

Significance

Returns
excessnumpy.ndarray

Excess