cstat

gammapy.stats.cstat(n_on, mu_on, n_on_min=1e-25)

C statistic, for Poisson data.

The C statistic is defined as

\[C = 2 \left[ \mu_{on} - n_{on} + n_{on} (\log(n_{on}) - log(\mu_{on}) \right]\]

and \(C = 0\) where \(\mu_{on} <= 0\).

n_on_min handles the case where n_on is 0 or less and the log cannot be taken. For more information see Fit statistics

Parameters:

n_on : array_like

Observed counts

mu_on : array_like

Expected counts

n_on_min : array_like

n_on = n_on_min where n_on <= n_on_min.

Returns:

stat : ndarray

Statistic per bin

References

• Sherpa stats page section on the C statistic
• Sherpa help page on the C statistic
• Cash 1979, ApJ 228, 939