Modeling causes of death: an integrated approach using CODEm

Table 7 Predictive validity measures for CODEm results for several selected causes of death

Cause	Model	RMSE (in-sample)	RMSE (test 2)	Trend test (in-sample)	Trend test (test 2)	Coverage (in-sample)	Coverage (test 2)
Cardiovascular disease (male)	Ensemble	0.123	0.290	0.764	0.699	0.993	0.853
	Best component model	0.122	0.292	0.764	0.693	0.992	0.822
Cardiovascular disease (female)	Ensemble	0.162	0.304	0.750	0.698	0.993	0.903
	Best component model	0.162	0.305	0.751	0.707	0.993	0.877
Maternal mortality	Ensemble	0.368	0.626	0.739	0.665	0.992	0.973
	Best component model	0.369	0.636	0.737	0.663	0.992	0.969
Cervical cancer	Ensemble	0.528	0.944	0.667	0.593	0.988	0.938
	Best component model	0.489	0.951	0.678	0.587	0.982	0.912
Breast cancer (female)	Ensemble	0.438	1.105	0.678	0.572	0.985	0.923
	Best component model	0.428	1.129	0.675	0.570	0.981	0.875
Lung cancer (male)	Ensemble	0.359	0.769	0.714	0.599	0.993	0.841
	Best component model	0.330	0.759	0.738	0.617	0.993	0.800
Lung cancer (female)	Ensemble	0.383	0.734	0.705	0.580	0.995	0.932
	Best component model	0.371	0.756	0.738	0.597	0.994	0.897

ISSN: 1478-7954