Skip to main content

Table 1 Diagnostic accuracy measures from the default and AIC-optimal models in logistic regression

From: Application of multiple statistical tests to enhance mass spectrometry-based biomarker discovery

Logistic Regression Model

Default

AIC-Optimal

Number of variables in final model

2

5

Goodness of Fit

0.669

0.882

AIC Statistic

69.65

57.80

Area under ROC curve

0.793

0.910

Sensitivity (%)

63.16

57.89

Specificity (%)

82.22

95.56

PPV (%)

85.96

84.62

NPV (%)

84.09

84.31

Percent accuracy (%)

76.56

84.38

  1. AIC = Akaike Information Criterion, ROC = Receiver Operating Characteristic, PPV = Positive Predictive Value, NPV = Negative Predictive Value