Table 3.

Temporal-longitudinal scenario: comparison between SESglmm and glmmLasso based on 20 replications with different target variable (gene) and independently randomly selected 2000 genes as predictor variables

Dataset	GDS5088	GDS4395	GDS4822	GDS3326	GDS3181	GDS4258	GDS3432	GDS3915
Average difference	-3.560(4.118)	0.188(0.516)	-0.003(0.134)	-0.180(0.506)	-0.020(0.04)	-0.139(0.288)	0.000(0.355)	0.093(0.455)
Proportion	19/20	7/20	9/20	13/20	15/20	10/20	10/20	8/20
p-value	0.0001 a	0.128	0.938	0.0015 a	0.0312 b	0.024 b	0.9946	0.3842

Average difference in performances (standard deviation of the differences appear inside the parentheses) and percentage of times SESglmm outperformed glmmLasso. The last line contains the permutation based p-value for the equality of the mean performances. Symbols ^a and ^b denote average differences that are statistically significant at 0.01 and 0.05, respectively. Notice that SESglmm is either statistically significantly better or on par with glmmLasso in terms of predictive performance