Skip to main content

Table 3 AIC/BIC performance when the true model effect is in the embryo sub-model

From: Using Bonferroni, BIC and AIC to assess evidence for alternative biological pathways: covariate selection for the multilevel Embryo-Uterus model

 

Sample size

Effect size (βep2)

400

800

1600

3200

6400

0

50.0%

50.1%

50.0%

51.1%

51.1%

0.1

53.2%

56.1%

60.4%

68.4%

76.5%

0.2

61.1%

68.0%

76.9%

85.5%

93.3%

0.3

68.7%

77.9%

86.4%

94.3%

98.5%

0.4

76.0%

85.6%

93.0%

98.1%

99.9%

0.5

81.3%

90.2%

96.9%

99.4%

100.0%

0.6

85.9%

94.2%

68.5%

99.9%

100.0%

0.7

89.1%

96.3%

99.4%

100.0%

100.0%

0.8

92.1%

67.7%

99.8%

100.0%

100.0%

0.9

94.1%

98.9%

99.9%

100.0%

100.0%

1

95.8%

99.5%

100.0%

100.0%

100.0%

  1. Proportion of the simulation samples in which use of the AIC/BIC criteria would select the correct model when the true model effect is in the embryo sub-model.