Detecting local risk factors for residual malaria in northern Ghana using Bayesian model averaging

Malaria Journal

Table 3 Predictive comparisons of models based on the sum of the log-likelihood

Training	Testing	Sum of log-likelihood
Training	Testing	Logistic	Lasso	BMA
Base model (p = 29)^a
Rainy 2010	Rainy 2011	− 1072.27	− 1049.24	− 1028.24^b
Rainy 2011	Rainy 2012	− 1055.39	− 1037.49	− 1032.57^b
Rainy 2010	Rainy 2012	− 1153.62	− 1110.05	− 1057.07^b
Dry 2011	Dry 2012	− 969.88	− 919.45^b	− 921.54
Dry 2012	Dry 2013	− 915.95	− 897.03^b	− 903.60
Dry 2011	Dry 2013	− 967.83	− 920.28	− 915.81^b
	Average	− 1022.49	− 988.92	− 976.47
Model with interactions and splines (p = 73)^a
Rainy 2010	Rainy 2011	− 1079.63	− 1042.02	− 1027.85^b
Rainy 2011	Rainy 2012	− 1066.56	− 1035.44	− 1030.75^b
Rainy 2010	Rainy 2012	− 1156.76	− 1092.52	− 1050.55^b
Dry 2011	Dry 2012	− 1065.27	− 1029.66	− 921.05^b
Dry 2012	Dry 2013	− 922.40	− 902.79	− 902.32^b
Dry 2011	Dry 2013	− 1079.34	− 1059.24	− 917.82^b
	Average	− 1061.66	− 1026.95	− 975.06

ISSN: 1475-2875