LikelihoodRatioResults

class sherpa.sim.simulate.LikelihoodRatioResults(ratios, stats, samples, lr, ppp, null, alt, parnames, parvals)[source] [edit on github]

Bases: NoNewAttributesAfterInit

The results of a likelihood ratio comparison simulation.

Changed in version 4.15.1: The parnames and parvals attributes have been added. They are intended to debug problem cases and so are not displayed by default.

ratios

The likelihood ratio for each simulation.

Type:: numpy array

stats

The fit statistic for the null and alternative models for each simulation. The shape is (nsim, 2).

Type:: numpy array

samples

The parameter samples array for each simulation, with shape (nsim, npar).

Type:: numpy array

lr

The likelihood ratio of the observed data for the null and alternate models.

Type:: number

ppp

The p value of the observed data for the null and alternate models.

Type:: number

null

The fit statistic of the null model on the observed data.

Type:: number

alt

The fit statistic of the alternate model on the observed data.

Type:: number

parnames

The thawed parameters in the alternate model.

Type:: list of str

parvals

The parameter values for each iteration for the alternate model, matching the order of parnames. The shape is (nsim, len(parnames)).

Type:: ndarray

Notes

The parvals field is useful to check that the simulations have not got stuck with certain parameter sets, for instance if the ratio value drops to ~ 0 and stays there. If this is the case then the analysis can be re-run after adjusting the range (the min or max range) of the parameters in question.

Methods Summary

format()

Convert the object to a string representation for display purposes.

Methods Documentation

format()[source] [edit on github]

Convert the object to a string representation for display purposes.

Returns:: txt – A string representation of the data stored in the object.
Return type:: str