Table 3.

Timings (seconds) for logistic model with lasso penalty and sparse features (95% zero). Total time for ten-fold cross-validation over a grid of 100 λ values.

	Logistic Regression — Sparse Features
	Correlation
	0	0.1	0.2	0.5	0.9	0.95
	N = 1000, p = 100
glmnet	0.77	0.74	0.72	0.73	0.84	0.88
l1lognet	5.19	5.21	5.14	5.40	6.14	6.26
BBR	2.01	1.95	1.98	2.06	2.73	2.88
	N = 100, p = 1000
glmnet	1.81	1.73	1.55	1.70	1.63	1.55
l1lognet	7.67	7.72	7.64	9.04	9.81	9.40
BBR	4.66	4.58	4.68	5.15	5.78	5.53
	N = 10, 000, p = 100
glmnet	3.21	3.02	2.95	3.25	4.58	5.08
l1lognet	45.87	46.63	44.33	43.99	45.60	43.16
BBR	11.80	11.64	11.58	13.30	12.46	11.83
	N = 100, p = 10, 000
glmnet	10.18	10.35	9.93	10.04	9.02	8.91
l1lognet	130.27	124.88	124.18	129.84	137.21	159.54
BBR	45.72	47.50	47.46	48.49	56.29	60.21