Skip to main content
. 2019 Mar 12;9:4229. doi: 10.1038/s41598-019-40450-8

Table 3.

Log-likelihood (LL), Deviance Information Criteria (DIC) and effective parameters (pD) for top models of mean BT incidence driven by a combination of host and landscape predictors.

Fixed effects in model LL DIC pD ∆ DIC Log-score RMSE Proportion of marginal variance explained by spatial effect ϕ
mean s.d.
bt ~ buffalos + nsheep + esheep −103.74 198.54 45.49 0.00 1.63 0.656 0.78 0.18
bt ~ irrig. crop + buffalos + nsheep + esheep −107.65 198.65 45.63 0.11 1.64 0.655 0.74 0.19
bt ~ irrig. crop + rain crop + buffalos + nsheep + esheep −111.99 198.87 45.95 0.33 1.65 0.650 0.73 0.19
bt ~ rain crop + buffalos + nsheep + esheep −108.35 198.93 45.97 0.39 1.65 0.647 0.75 0.18
bt ~ irrig. crop + rain crop + buffalos + esheep −108.38 199.22 45.41 0.68 1.63 0.664 0.77 0.17
bt ~ rain crop + buffalos + esheep −105.33 199.94 45.72 1.40 1.64 0.658 0.79 0.16
bt ~ irrig. crop + rain crop + nsheep + esheep −108.97 200.28 46.47 1.74 1.67 0.653 0.68 0.20
bt ~ irrig. crop + rain crop + buffalos + nsheep −109.06 200.32 46.20 1.78 1.65 0.657 0.77 0.17
bt ~ irrig. crop + rain crop + buffalos −105.19 200.37 45.64 1.83 1.64 0.669 0.79 0.15
* bt ~ 1 −103.81 210.52 49.11 11.98 1.70 0.625 0.87 0.11

The null intercept-only model* without any covariates is given at the bottom for comparison.