Bayesian data analysis for fusion diagnostics

The electron density and the electron temperature profiles are reconstructed from ruby-laserbased Thomson scattering data on Wendelstein 7-AS (W7-AS) by using Bayesian probability theory (BPT). An outlier identification method based on Bayesian statistics is applied to least-squares minimization fits in order to remove unwanted influences on the electron density and the electron temperature profiles. All uncertainties of the measured data, the calibration measurements, the physical model parameters, and the measurement nuisance parameters are considered in the systematic error analysis. We compare our results with the results from a conventional analysis to determine the electron temperature and density profiles from Thornson scattering data on W7-AS.