12.3 Normal Regression

# Simulate the Normal model
equality_normal_sim <- stan_glm(laws ~ percent_urban + historical, 
                                data = equality, 
                                family = gaussian,
                                prior_intercept = normal(7, 1.5),
                                prior = normal(0, 2.5, autoscale = TRUE),
                                prior_aux = exponential(1, autoscale = TRUE),
                                chains = 4, iter = 5000*2, seed = 84735)

12.3.1 Posterior Predictive Check

pp_check(equality_normal_sim, plotfun = "hist", nreps = 5) + 
  geom_vline(xintercept = 0) + 
  xlab("laws")

• observe that some of the predicted counts (for number of laws) are negative!