Synopsis
The CStat statistic adjusted to penalize negative model values
Description
The cstatnegativepenalty statistic is the same as the cstat statistic except when the model predicts one or more negative values. In this case the statistic is penalised by the absolute value of the negative value, or values. The aim is to help the optimiser out in moving the fit to a more-physical location in the search space.
Example
sherpa> set_stat("cstatnegativepenalty")
sherpa> get_stat_name()
'cstatnegativepenalty'Set the fitting statistic and then confirm the new value.
Changes in CIAO 4.18
This statistic is new in CIAO 4.18.
Bugs
See the bugs pages on the Sherpa website for an up-to-date listing of known bugs.
See Also
- info
- list_stats
- statistics
- cash, chi2constvar, chi2datavar, chi2gehrels, chi2modvar, chi2xspecvar, chisquare, cstat, get_prior, leastsq, list_priors, set_prior, set_sampler, set_stat, wstat