This repository contains the R code to reproduce the analyses of the paper entitled "A proposal for homoskedastic modelling with conditional auto-regressive distributions" (Martínez-Beneito et al., 2025).
The results of the models presented in this paper were obtained using the R-INLA stable version 24.06.27 on R 4.4.1.
This folder contains the following files:
-
An .Rdata file with
sfobjects containing the cartographies of municipalities in the autonomous regions of the Valencian region (CartoVR), Castile and Leon (CartoCL), Aragon (CartoAR) and Castile-La Mancha (CartoCM). -
An .Rdata file with the posterior mean estimates of log-risks for 100 causes of death across municipalities in the Valencian region (
Res.VR), Castile and Leon (Res.CL), Aragon (Res.AR) and Castile-La Mancha (Res.CM), obtained from models fitted using the traditional ICAR prior. -
An .Rdata file with the posterior mean estimates of log-risks for 100 causes of death across municipalities in the Valencian region (
Res.VR.HomCAR), Castile and Leon (Res.CL.HomCAR), Aragon (Res.AR.HomCAR) and Castile-La Mancha (Res.CM.HomCAR), obtained from models fitted using the proposed HomCAR prior. -
R script to reproduce the results from Section 2 (The ICAR distribution and its marginal variances) of the paper, computing the marginal variances of the ICAR distribution for municipalities in four autonomous regions of Spain and comparing them with the empirical variances of the posterior mean log-risk estimates across 100 causes of death analyzed in each region.
-
R script to reproduce the results from Section 4 (A practical assessment of the HomCAR distribution) of the paper, presenting a simulation study that compares the performance of the proposed HomCAR distribution with the traditional ICAR prior.
This research was supported by the projects PID2020-113125RBI00/MCIN/AEI/10.13039/501100011033, PID2022-136455NBI00/MCIN/AEI/10.13039/501100011033 and PID2024-155382OBI00/MICIU/AEI/10.13039/501100011033 funded by the Ministry of Science, Innovation and Universities of Spain and FEDER (UE) and grant CIAICO PID2022-136455NB-I00 funded by Dirección General de Ciencia e Investigación (Generalitat Valenciana).
