Supporting Bayesian Analysis

[JanWeldert]

At the moment we focus on the frequentist approach. To also support Bayesian analyses we would mainly need some sort of MCMC sampling tool and its integration into the analysis class. Examples how to perform a Bayesian analysis would also be useful.

[JoshHenryPeterson]

I talked with some MSU people and they pointed me towards some work on implementing MCMC. Apparently they were able to generate a sensitivity? Their code lives in the fridge here: analysis/DRAGON_NSI_3y/scripts/sampling. It's a bit old so there may be some technical debt but it could be a good starting point.