Skip to main content

Table 3 Stepwise and Consensus Classification Models

From: Cytokine expression profiles of immune imbalance in post-mononucleosis chronic fatigue

Classification based on

Stepwise model of IL-1a, 6, 8,13, 23

Consensus model of IL-6, 8, and 23

Minimal model of IL-6 and 8

Correct Rate

0.93

0.86

0.79

Error Rate

0.07

0.14

0.21

Inconclusive Rate

0.00

0.00

0.00

Classified Rate

1.00

1.00

1.00

Sensitivity

0.92

0.83

0.79

Specificity

0.94

0.89

0.78

Positive Predictive Value

0.96

0.91

0.83

Negative Predictive Value

0.89

0.80

0.74

Positive Likelihood

16.50

7.50

3.56

Negative Likelihood

0.09

0.19

0.27

  1. Performance of linear classification models based on (i) a stepwise selection of IL-1a, 6, 8, 13 and 23, (ii) the intersecting subset of IL-6, 8 and 23 found in both stepwise and all-possible-subsets models and (iii) a core selection of IL-6 and 8, the two most robust choices based on simulations with random data substitutions.