Table 4.

Performance summary of random forest models on simulated threenorm data sets with 20 relevant predictors and k noise predictors. Performance metrics were computed using independent test sets of size 1000 (Test) and the out-of-bag data (OOB), and averaged over 20 simulation runs. PCC is the percent of observations correctly classified, sensitivity is the percent of observations in class 1 correctly classified, specificity is the percent of observations in class 2 correctly classified, and AUC is the area under the receiver operating characteristic curve

	k	PCC	Sens.	Spec.	AUC
Test	0	0.857	0.854	0.861	0.936
OOB	0	0.862	0.862	0.862	0.937
Test	50	0.849	0.841	0.858	0.929
OOB	50	0.848	0.832	0.864	0.927
Test	100	0.844	0.835	0.853	0.926
OOB	100	0.847	0.842	0.850	0.925
Test	150	0.843	0.835	0.852	0.924
OOB	150	0.833	0.827	0.838	0.915
Test	200	0.837	0.831	0.843	0.920
OOB	200	0.834	0.827	0.840	0.912
Test	250	0.836	0.843	0.831	0.917
OOB	250	0.829	0.833	0.825	0.909
Test	300	0.834	0.827	0.842	0.918
OOB	300	0.828	0.822	0.832	0.909