Modeling the diagnosis of coronary artery disease by discriminant analysis and logistic regression: a cross-sectional study

Table 5 Differences in surface area under the ROC curve in linear discriminant analysis, quadratic discriminant analysis, K nearest neighbor and logistic regression

Two by two comparison	Difference between areas	Standard error	95% CI	z statistic	P-value
KNN
LDA	0.0009	0.011	(− 0.021 to 0.02)	0.782	0.93
LR	0.00003	0.011	(− 0.023 to 0.023)	0.002	0.99
QDA	0.01	0.014	(− 0.017 to 0.039)	0.73	0.46
LDA
LR	0.0008	0.001	(− 0.001 to 0.0036)	0.61	0.54
QDA	0.009	0.014	(− 0.01 to 0.03)	0.69	0.49
LR
QDA	0.01	0.01	(− 0.016 to 0.037)	0.75	0.44

LDA linear discriminant analysis, QDA quadratic discriminant analysis, KNN K-nearest neighbor, LR logistic regression

ISSN: 1472-6947