Table 2.

Univariate analysis: DIC values for various models under first- and last-activation schemes and spatial frailties in latent link η or cure fraction parameter θ with CAR assumption

Model	Univariate: CAR assumption	First		Last
	Spatial frailty in η	pD	DIC	pD	DIC
1(a)	${\eta }_{ij}={\mathbf{x}}_{ij}^{\mathrm{T}}\beta +{\varphi }_i;{\theta }_{ij}=\theta ;\mathit{N}~\text{Po}(\theta )$	23.406	3629.240	23.310	3628.890
1(b)	${\eta }_{ij}={\mathbf{x}}_{ij}^{\mathrm{T}}\beta +{\varphi }_i;{\theta }_{ij}=\theta ;\mathit{N}~\text{Ber}(\theta )$	20.828	3628.400	20.828	3628.400
1(c)	${\eta }_{ij}={\mathbf{x}}_{ij}^{\mathrm{T}}\beta +{\varphi }_i;{\theta }_{ij}=\theta ;\mathit{N}~\text{Geo}(\theta )$	15.820	3621.870	23.454	3627.090
2	$\begin{array}{l}\text{log}({\theta }_{ij})={\mathbf{x}}_{ij}^{\mathrm{T}}\beta ;{\eta }_{ij}={\eta }_0+{\varphi }_i;\mathit{N}~\text{Po}(\theta )\hfill \\ \text{Spatial}\text{frailty}\text{in}\theta \hfill \end{array}$	5.223	3630.930	3.566	3634.050
1(a)	${\eta }_{ij}={\mathbf{x}}_{ij}^{\mathrm{T}}\beta ;\text{log}({\theta }_{ij})={\theta }_0+{\varphi }_i;\mathit{N}~\text{Po}(\theta )$	6.011	3632.200	6.343	3636.690
2	$\text{log}({\theta }_{ij})={\mathbf{x}}_{ij}^{\mathrm{T}}\beta +{\varphi }_i;{\eta }_{ij}=\eta ;\mathit{N}~\text{Po}(\theta )$	14.850	3630.070	12.656	3640.040