Joint Bayesian modeling of time to malaria and mosquito abundance in Ethiopia

BMC Infectious Diseases

Table 3 Parameter estimates and 95% credible intervals for the joint model

	Parameter	Posterior mean	2.5%	97.5%
Abundance model
Intercept	β ₀	0.18	-0.03	0.39
S ₁(r a i n(t))	β ₁	4.37	4.16	4.59
S ₂(r a i n(t))	β ₂	0.92	0.85	0.98
Measurement error	σ	0.71	0.70	0.71
Time to event model
Age	θ ₁	0.01	-0.03	0.05
Gender	θ ₂	-0.05	-0.22	0.12
Association main effect	α ₁	0.12	0.04	0.19
Association interaction	α ₂	0.26	0.16	0.36
Hyper-parameters
Penalty	λ	0.003	0.002	0.006
Random effect covariance	D _1,1	23.55	22.15	25.05
Random effect covariance	D _2,1	0.58	0.26	0.89
Random effect covariance	D _3,1	-0.73	-0.99	-0.49
Random effect covariance	D _2,2	2.16	2.01	2.31
Random effect covariance	D _3,2	0.81	0.72	0.91
Random effect covariance	D _3,3	1.42	1.32	1.52
DIC	403452.2

D _i,j denotes the ij-element of the covariance matrix for the random effects. We use a three week window to define the incidence I _k(i)(t). In this run, the distance is not included in the model

ISSN: 1471-2334