A copula-based quantifying of the relationship between race inequality among neighbourhoods in São Paulo and age at death

Open Access

Table 2

Summaries of the Bayesian estimation.

copula	prior	mean	s-m	sd	2.5%	25%	50%	75%	97.5%	n_eff	Rhat
Frank	NI	−23.936	0.062	2.185	−28.396	−25.384	−23.854	−22.440	−19.903	1255	1.006
Frank	I	−24.108	0.061	2.139	−28.376	−25.593	−24.057	−22.614	−20.035	1216	1.001
Gaussian	NI	−0.953	0.000	0.007	−0.965	−0.957	−0.953	−0.948	−0.937	1458	1.000
Gaussian	I	−0.953	0.000	0.007	−0.965	−0.958	−0.954	−0.949	−0.937	1210	1.002

n_eff: final number of simulations used for the estimation; sd: standard deviation; s–m=sd/n_eff^1/2; Rhat: potential scale reduction factor on split chains (at convergence, Rhat = 1). In bold letter the Bayesian estimates of θ for Frank and ρ for Gaussian copula, by quadratic loss function on left, by multi linear loss function on right. Non-Informative (NI) prior on top and Informative (I) prior on bottom.