Used correct value of PET coefficient and redid all the Bayesian inference. This changed the inferred value of bulk loss angle little bit.