To create PARIS outputs, we need samples from the prior, prior predictive, and posterior predictive distributions.
Also, the Y trace currently created by hbmcmc_postprocess doesn't take into account model and observation error, so the posterior Y uncertainties currently reported are lower than they should be.
To create PARIS outputs, we need samples from the prior, prior predictive, and posterior predictive distributions.
Also, the Y trace currently created by
hbmcmc_postprocess
doesn't take into account model and observation error, so the posterior Y uncertainties currently reported are lower than they should be.