SuperNova statistical analysis package (sn_stat)¶
sn_stat provides the implementation of the statistical methods used for search for the supernova neutrino signal in the series of neutrino interactions.
Task¶
Experiment measures the neutrino interactions in the detector, recording their timestamps \(\{t_i\}\).
The statistical analysis needs to determine if the measured interactions agree with the background rate, or if they contain a signal from the supernova, and, potentially, to identify the time of the signal start.
When there is no supernova, the expected event rate at time t is \(B(t)\) events/s. In case of supernova starting at time \(t_0\), the event rate vs. time is \(B(t)+S(t-t_0)\).
The signal significance value is calculated for every assumed supernova time \(t_0\), measuring the deviation from the background-only hypothesis.
What sn_stat can do:¶
Define and manipulate the event rates vs. time
sn_stat.rate()Calculate the Log Likelihood Ratio
sn_stat.LLRCalculate the significance of the supernova signal observation in the measured data
sn_stat.ShapeAnalysis,:class:sn_stat.CountAnalysisAnd the expected distribution of the significance for a given hypothesis (rate)
Generate the events timestamps based on the event rate
sn_stat.Sampler
Contents: