Inference for Bugs model at "modelBPBM.txt", fit using jags, 3 chains, each with 2000 iterations (first 1000 discarded), n.thin = 10 n.sims = 300 iterations saved mu.vect sd.vect 2.5% 25% 50% 75% 97.5% Rhat a[1,1] 1.809 0.571 0.629 1.456 1.849 2.246 2.771 1.005 a[2,1] 1.525 0.578 0.384 1.117 1.588 1.986 2.452 1.017 a[3,1] -0.121 0.459 -1.103 -0.403 -0.048 0.121 0.746 1.010 a[4,1] 1.396 0.608 0.132 1.015 1.463 1.840 2.412 1.004 a[5,1] 1.615 0.587 0.445 1.199 1.625 2.058 2.568 1.019 a[1,2] -0.135 0.604 -1.375 -0.465 -0.083 0.165 1.141 1.017 a[2,2] 0.183 0.571 -1.022 -0.119 0.132 0.514 1.299 1.007 a[3,2] 0.317 0.647 -0.953 -0.097 0.264 0.752 1.562 1.058 a[4,2] 0.115 0.675 -1.336 -0.259 0.087 0.455 1.510 0.998 a[5,2] -0.325 0.742 -2.104 -0.607 -0.218 0.091 0.987 1.004 a[1,3] -0.238 1.011 -2.171 -0.761 -0.240 0.238 1.894 1.076 a[2,3] 1.360 1.128 -0.454 0.542 1.236 2.000 4.134 1.023 a[3,3] 0.219 1.134 -2.097 -0.358 0.058 0.776 2.762 1.001 a[4,3] -0.086 1.016 -2.098 -0.528 -0.075 0.395 2.230 1.033 a[5,3] 0.593 1.053 -1.122 -0.065 0.406 1.169 3.121 1.091 sdgamma[1,1] 2.624 1.164 0.760 1.746 2.468 3.437 4.837 1.003 sdgamma[2,1] 2.636 1.258 0.663 1.537 2.525 3.656 4.916 0.998 sdgamma[3,1] 1.450 1.364 0.054 0.388 0.890 2.254 4.477 0.998 sdgamma[4,1] 2.495 1.255 0.509 1.448 2.367 3.505 4.815 1.000 sdgamma[5,1] 2.690 1.241 0.566 1.770 2.529 3.711 4.907 1.002 sdgamma[1,2] 1.594 1.359 0.072 0.460 1.116 2.464 4.805 0.997 sdgamma[2,2] 1.715 1.338 0.115 0.579 1.301 2.542 4.700 1.005 sdgamma[3,2] 1.804 1.292 0.069 0.734 1.531 2.723 4.650 1.006 sdgamma[4,2] 1.714 1.333 0.068 0.571 1.398 2.520 4.577 1.054 sdgamma[5,2] 1.763 1.373 0.083 0.599 1.402 2.692 4.541 1.006 sdgamma[1,3] 1.954 1.356 0.039 0.755 1.809 2.947 4.637 1.005 sdgamma[2,3] 2.421 1.349 0.284 1.331 2.204 3.588 4.795 1.025 sdgamma[3,3] 2.039 1.363 0.087 0.914 1.860 3.091 4.718 1.014 sdgamma[4,3] 1.629 1.321 0.073 0.551 1.227 2.463 4.553 1.002 sdgamma[5,3] 2.131 1.450 0.099 0.828 1.890 3.239 4.868 1.004 deviance -57.393 7.528 -69.966 -63.022 -58.038 -52.114 -42.156 0.999 n.eff a[1,1] 240 a[2,1] 290 a[3,1] 300 a[4,1] 300 a[5,1] 110 a[1,2] 140 a[2,2] 190 a[3,2] 37 a[4,2] 300 a[5,2] 300 a[1,3] 38 a[2,3] 110 a[3,3] 300 a[4,3] 170 a[5,3] 34 sdgamma[1,1] 300 sdgamma[2,1] 300 sdgamma[3,1] 300 sdgamma[4,1] 300 sdgamma[5,1] 300 sdgamma[1,2] 300 sdgamma[2,2] 290 sdgamma[3,2] 300 sdgamma[4,2] 110 sdgamma[5,2] 300 sdgamma[1,3] 300 sdgamma[2,3] 120 sdgamma[3,3] 150 sdgamma[4,3] 300 sdgamma[5,3] 300 deviance 300 For each parameter, n.eff is a crude measure of effective sample size, and Rhat is the potential scale reduction factor (at convergence, Rhat=1). DIC info (using the rule, pD = var(deviance)/2) pD = 28.5 and DIC = -28.9 DIC is an estimate of expected predictive error (lower deviance is better).