Speaker
Oliver Schulz
(MPI for Physics, Munich)
Description
GERDA is an ultra-low background experiment, designed to search for the
neutrinoless double beta decay of Ge-76. The main background sources of
such an experiment are minute radioactive ontaminations of the experimental
setup itself. Gaining a good understanding of the individual contributions
to this radioactive background is vital not only for data analysis, but also
guides the design of the next stage of the experiment.
The Bayesian Analysis Toolkit (BAT) was used to perform a full background
decomposition of the GERDA Phase-I data. The Bayesian approach allowed the
implementation of prior knowledge and the ability to handle competing models
in a consistent way. It also yields a straightforward uncertainty propagation,
taking the correlations between model components into account.
We describe the formulation of the analysis problem and the technical
realization. The techniques described here are of general validity and interest,
and have proven to be very successful.
Primary author
Oliver Schulz
(MPI for Physics, Munich)
