Skip to main content

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

From: Feature selection for high-dimensional temporal data

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
  1. 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