model{ for (i in 1:261){ xb.corrup[i] <- gamma.corrup[1] + gamma.corrup[2]*lag_corrup[i]+ gamma.corrup[3]*polity2[i] + gamma.corrup[4]*frac_reported[i] + gamma.corrup[5]*polity2[i]*frac_reported[i] + gamma.corrup[6]*rgdpch[i] + gamma.corrup[7]*newscirc[i] + gamma.corrup[8]*newscirc[i]*polity2[i] corrup[i] ~ dnorm(xb.corrup[i], tau.corrup) xb.laword[i] <- gamma.laword[1] + gamma.laword[2]*lag_laword[i]+ gamma.laword[3]*polity2[i] + gamma.laword[4]*frac_reported[i] + gamma.laword[5]*polity2[i]*frac_reported[i] + gamma.laword[6]*rgdpch[i]+ gamma.laword[7]*newscirc[i] + gamma.laword[8]*newscirc[i]*polity2[i] laword[i] ~ dnorm(xb.laword[i], tau.laword) xb.burqual[i] <- gamma.burqual[1] + gamma.burqual[2]*lag_burqual[i]+ gamma.burqual[3]*polity2[i] + gamma.burqual[4]*frac_reported[i] + gamma.burqual[5]*polity2[i]*frac_reported[i] + gamma.burqual[6]*rgdpch[i]+ gamma.burqual[7]*newscirc[i] + gamma.burqual[8]*newscirc[i]*polity2[i] burqual[i] ~ dnorm(xb.burqual[i], tau.burqual) } #Specify priors for the regression model #prior means g[1] <- 0 g[2] <- 0 g[3] <- 0 g[4] <- 0 g[5] <- 0 g[6] <- 0 g[7] <-0 g[8] <- 0 #prior variance covariance for (i in 1:7){ s.g[i,i]<- .001 for (j in (i+1):8){ s.g[i,j] <- 0 s.g[j,i]<-0 } } s.g[8,8]<-.001 sigma.corrup ~ dunif(0,100) tau.corrup <- pow(sigma.corrup,-2) sigma.laword ~ dunif(0,100) tau.laword <- pow(sigma.laword,-2) sigma.burqual ~ dunif(0,100) tau.burqual <- pow(sigma.burqual,-2) gamma.corrup[1:8] ~ dmnorm(g[1:8], s.g[1:8, 1:8]) gamma.laword[1:8] ~ dmnorm(g[1:8], s.g[1:8, 1:8]) gamma.burqual[1:8] ~ dmnorm(g[1:8], s.g[1:8, 1:8]) }