Representation of the relations between parameters and observations in our RV model, as a probabilistic graphical model. An arrow between two nodes indicates the direction of conditional dependence. The circled nodes are the parameters of the model, whose joint distribution is sampled by the Markov chain Monte Carlo (MCMC) algorithm. The double circled node v_i represents the observed RVs. The filled nodes represent deterministic variables: if these variables have parent nodes ( and Σ), they are given by a deterministic function of those parents; if they do not have parents (t_i and σ_i), they are assumed as being given and thus fixed. The variables inside boxes are repeated a number of times, as shown in the bottom left corner of each box.