Table 2.

Average out-of-sample test errors (rounded average model sparsity in parenthesis) for trac ($a\in \{1,1/2\}$) and sparse log-contrast models, respectively. Each row represents the results for base level OTU, genus, and family. Each value is averaged over ten different training/test splits of the Central Park soil data.

Base level	p	trac (a $=$ 1)	trac (a $=$ 1/2)	Sparse log-contrast
OTU	3379	0.40 (10)	0.39 (18)	0.39 (33)
Genus	2779	0.40 (13)	0.38 (22)	0.39 (26)
Family	1492	0.39 (10)	0.39 (15)	0.40 (29)