Skip to main content

Table 3 Comparison of estimated time and conditional AIC between indiCAR and other methods when data are generated without spatial random effect, \({\varvec{\lambda }}=0\)

From: Individual level covariate adjusted conditional autoregressive (indiCAR) model for disease mapping

Sample per group Total sample Time to convergence (s) Conditional AIC
indiCAR glmer with random intercept hlmer with random intercept hlmer with CAR indiCAR glmer with random intercept hlmer with random intercept hlmer with CAR
Data generated in 100 groups
 1:50 2373 0.73 1.98 0.43 2.36 1419.26 1492.26 1445.65 1445.87
 1:100 5056 2.09 5.26 0.55 2.93 3170.9 3225.28 3194.03 3193.98
 1:500 26,473 10.34 23.63 1.65 12.96 15,996.05 15,968.02 15,955.41 15,955.40
 1:1000 48,778 37.25 53.25 3.01 29.68 31,063.34 31,011.67 31,001.58 31,001.72
Data generated in 400 groups
 1:50 10,192 51.39 9.44 2.64 97.45 6027.28 6242.24 6097.72 6097.84
 1:100 19,843 73.31 33.13 11.39 244.02 12,017.28 12,185.15 12,037.20 12037.27
 1:500 98,870 140.74 71.96 38.91 Not feasible 59,061.39 58,929.01 58,879.30 Not feasible
 1:1000 205,952 207.84 214.51 149.96 Not feasible 121,733.50 121,533.80 121,510.20 Not feasible