DeLong pairwise AUC tests
==========================

RF vs LR

	DeLong's test for two correlated ROC curves

data:  model_rocs[[i]] and model_rocs[[j]]
Z = -0.5498, p-value = 0.5825
alternative hypothesis: true difference in AUC is not equal to 0
95 percent confidence interval:
 -0.03999296  0.02247086
sample estimates:
AUC of roc1 AUC of roc2 
  0.7032934   0.7109709 
 

RF vs SVM

	DeLong's test for two correlated ROC curves

data:  model_rocs[[i]] and model_rocs[[j]]
Z = 2.938, p-value = 0.003303
alternative hypothesis: true difference in AUC is not equal to 0
95 percent confidence interval:
 0.01612303 0.08074038
sample estimates:
AUC of roc1 AUC of roc2 
  0.7032934   0.6537782 
 

RF vs DT

	DeLong's test for two correlated ROC curves

data:  model_rocs[[i]] and model_rocs[[j]]
Z = 9.4627, p-value < 2.2e-16
alternative hypothesis: true difference in AUC is not equal to 0
95 percent confidence interval:
 0.2826381 0.4303075
sample estimates:
AUC of roc1 AUC of roc2 
  0.7032934   0.3457371 
 

RF vs NB

	DeLong's test for two correlated ROC curves

data:  model_rocs[[i]] and model_rocs[[j]]
Z = -0.060848, p-value = 0.9515
alternative hypothesis: true difference in AUC is not equal to 0
95 percent confidence interval:
 -0.03361777  0.03159325
sample estimates:
AUC of roc1 AUC of roc2 
  0.7032934   0.7032221 
 

RF vs KNN

	DeLong's test for two correlated ROC curves

data:  model_rocs[[i]] and model_rocs[[j]]
Z = 6.1598, p-value = 7.285e-10
alternative hypothesis: true difference in AUC is not equal to 0
95 percent confidence interval:
 0.1593618 0.3081032
sample estimates:
AUC of roc1 AUC of roc2 
  0.7032934   0.4684773 
 

RF vs XGB

	DeLong's test for two correlated ROC curves

data:  model_rocs[[i]] and model_rocs[[j]]
Z = 8.9637, p-value < 2.2e-16
alternative hypothesis: true difference in AUC is not equal to 0
95 percent confidence interval:
 0.1579956 0.2464242
sample estimates:
AUC of roc1 AUC of roc2 
  0.7032934   0.5000000 
 

RF vs ANN

	DeLong's test for two correlated ROC curves

data:  model_rocs[[i]] and model_rocs[[j]]
Z = -1.7605, p-value = 0.07832
alternative hypothesis: true difference in AUC is not equal to 0
95 percent confidence interval:
 -0.056914820  0.003051119
sample estimates:
AUC of roc1 AUC of roc2 
  0.7032934   0.7291417 
 

RF vs LDA

	DeLong's test for two correlated ROC curves

data:  model_rocs[[i]] and model_rocs[[j]]
Z = -1.3792, p-value = 0.1678
alternative hypothesis: true difference in AUC is not equal to 0
95 percent confidence interval:
 -0.050154368  0.008722946
sample estimates:
AUC of roc1 AUC of roc2 
  0.7032934   0.7229256 
 

RF vs QDA

	DeLong's test for two correlated ROC curves

data:  model_rocs[[i]] and model_rocs[[j]]
Z = 2.0518, p-value = 0.04019
alternative hypothesis: true difference in AUC is not equal to 0
95 percent confidence interval:
 0.001802332 0.078722333
sample estimates:
AUC of roc1 AUC of roc2 
  0.7032934   0.6619475 
 

LR vs SVM

	DeLong's test for two correlated ROC curves

data:  model_rocs[[i]] and model_rocs[[j]]
Z = 2.6895, p-value = 0.007155
alternative hypothesis: true difference in AUC is not equal to 0
95 percent confidence interval:
 0.01551405 0.09887147
sample estimates:
AUC of roc1 AUC of roc2 
  0.7109709   0.6537782 
 

LR vs DT

	DeLong's test for two correlated ROC curves

data:  model_rocs[[i]] and model_rocs[[j]]
Z = 9.3877, p-value < 2.2e-16
alternative hypothesis: true difference in AUC is not equal to 0
95 percent confidence interval:
 0.2889806 0.4414870
sample estimates:
AUC of roc1 AUC of roc2 
  0.7109709   0.3457371 
 

LR vs NB

	DeLong's test for two correlated ROC curves

data:  model_rocs[[i]] and model_rocs[[j]]
Z = 0.43456, p-value = 0.6639
alternative hypothesis: true difference in AUC is not equal to 0
95 percent confidence interval:
 -0.02720018  0.04269776
sample estimates:
AUC of roc1 AUC of roc2 
  0.7109709   0.7032221 
 

LR vs KNN

	DeLong's test for two correlated ROC curves

data:  model_rocs[[i]] and model_rocs[[j]]
Z = 6.8059, p-value = 1.004e-11
alternative hypothesis: true difference in AUC is not equal to 0
95 percent confidence interval:
 0.1726601 0.3123271
sample estimates:
AUC of roc1 AUC of roc2 
  0.7109709   0.4684773 
 

LR vs XGB

	DeLong's test for two correlated ROC curves

data:  model_rocs[[i]] and model_rocs[[j]]
Z = 9.4755, p-value < 2.2e-16
alternative hypothesis: true difference in AUC is not equal to 0
95 percent confidence interval:
 0.1673327 0.2546092
sample estimates:
AUC of roc1 AUC of roc2 
  0.7109709   0.5000000 
 

LR vs ANN

	DeLong's test for two correlated ROC curves

data:  model_rocs[[i]] and model_rocs[[j]]
Z = -1.9109, p-value = 0.05602
alternative hypothesis: true difference in AUC is not equal to 0
95 percent confidence interval:
 -0.0368084703  0.0004668678
sample estimates:
AUC of roc1 AUC of roc2 
  0.7109709   0.7291417 
 

LR vs LDA

	DeLong's test for two correlated ROC curves

data:  model_rocs[[i]] and model_rocs[[j]]
Z = -1.3292, p-value = 0.1838
alternative hypothesis: true difference in AUC is not equal to 0
95 percent confidence interval:
 -0.029582337  0.005673013
sample estimates:
AUC of roc1 AUC of roc2 
  0.7109709   0.7229256 
 

LR vs QDA

	DeLong's test for two correlated ROC curves

data:  model_rocs[[i]] and model_rocs[[j]]
Z = 2.6903, p-value = 0.007139
alternative hypothesis: true difference in AUC is not equal to 0
95 percent confidence interval:
 0.01330846 0.08473831
sample estimates:
AUC of roc1 AUC of roc2 
  0.7109709   0.6619475 
 

SVM vs DT

	DeLong's test for two correlated ROC curves

data:  model_rocs[[i]] and model_rocs[[j]]
Z = 8.1592, p-value = 3.374e-16
alternative hypothesis: true difference in AUC is not equal to 0
95 percent confidence interval:
 0.2340446 0.3820375
sample estimates:
AUC of roc1 AUC of roc2 
  0.6537782   0.3457371 
 

SVM vs NB

	DeLong's test for two correlated ROC curves

data:  model_rocs[[i]] and model_rocs[[j]]
Z = -2.2856, p-value = 0.02228
alternative hypothesis: true difference in AUC is not equal to 0
95 percent confidence interval:
 -0.091844020 -0.007043919
sample estimates:
AUC of roc1 AUC of roc2 
  0.6537782   0.7032221 
 

SVM vs KNN

	DeLong's test for two correlated ROC curves

data:  model_rocs[[i]] and model_rocs[[j]]
Z = 4.8351, p-value = 1.331e-06
alternative hypothesis: true difference in AUC is not equal to 0
95 percent confidence interval:
 0.1101864 0.2604153
sample estimates:
AUC of roc1 AUC of roc2 
  0.6537782   0.4684773 
 

SVM vs XGB

	DeLong's test for two correlated ROC curves

data:  model_rocs[[i]] and model_rocs[[j]]
Z = 6.2517, p-value = 4.06e-10
alternative hypothesis: true difference in AUC is not equal to 0
95 percent confidence interval:
 0.1055674 0.2019889
sample estimates:
AUC of roc1 AUC of roc2 
  0.6537782   0.5000000 
 

SVM vs ANN

	DeLong's test for two correlated ROC curves

data:  model_rocs[[i]] and model_rocs[[j]]
Z = -3.5804, p-value = 0.0003431
alternative hypothesis: true difference in AUC is not equal to 0
95 percent confidence interval:
 -0.11661879 -0.03410833
sample estimates:
AUC of roc1 AUC of roc2 
  0.6537782   0.7291417 
 

SVM vs LDA

	DeLong's test for two correlated ROC curves

data:  model_rocs[[i]] and model_rocs[[j]]
Z = -3.5837, p-value = 0.0003387
alternative hypothesis: true difference in AUC is not equal to 0
95 percent confidence interval:
 -0.10696463 -0.03133021
sample estimates:
AUC of roc1 AUC of roc2 
  0.6537782   0.7229256 
 

SVM vs QDA

	DeLong's test for two correlated ROC curves

data:  model_rocs[[i]] and model_rocs[[j]]
Z = -0.38586, p-value = 0.6996
alternative hypothesis: true difference in AUC is not equal to 0
95 percent confidence interval:
 -0.04966532  0.03332657
sample estimates:
AUC of roc1 AUC of roc2 
  0.6537782   0.6619475 
 

DT vs NB

	DeLong's test for two correlated ROC curves

data:  model_rocs[[i]] and model_rocs[[j]]
Z = -9.7452, p-value < 2.2e-16
alternative hypothesis: true difference in AUC is not equal to 0
95 percent confidence interval:
 -0.4293830 -0.2855871
sample estimates:
AUC of roc1 AUC of roc2 
  0.3457371   0.7032221 
 

DT vs KNN

	DeLong's test for two correlated ROC curves

data:  model_rocs[[i]] and model_rocs[[j]]
Z = -3.9474, p-value = 7.901e-05
alternative hypothesis: true difference in AUC is not equal to 0
95 percent confidence interval:
 -0.18368338 -0.06179708
sample estimates:
AUC of roc1 AUC of roc2 
  0.3457371   0.4684773 
 

DT vs XGB

	DeLong's test for two correlated ROC curves

data:  model_rocs[[i]] and model_rocs[[j]]
Z = -8.1417, p-value = 3.897e-16
alternative hypothesis: true difference in AUC is not equal to 0
95 percent confidence interval:
 -0.1913988 -0.1171270
sample estimates:
AUC of roc1 AUC of roc2 
  0.3457371   0.5000000 
 

DT vs ANN

	DeLong's test for two correlated ROC curves

data:  model_rocs[[i]] and model_rocs[[j]]
Z = -10.105, p-value < 2.2e-16
alternative hypothesis: true difference in AUC is not equal to 0
95 percent confidence interval:
 -0.4577681 -0.3090411
sample estimates:
AUC of roc1 AUC of roc2 
  0.3457371   0.7291417 
 

DT vs LDA

	DeLong's test for two correlated ROC curves

data:  model_rocs[[i]] and model_rocs[[j]]
Z = -9.8064, p-value < 2.2e-16
alternative hypothesis: true difference in AUC is not equal to 0
95 percent confidence interval:
 -0.4525757 -0.3018012
sample estimates:
AUC of roc1 AUC of roc2 
  0.3457371   0.7229256 
 

DT vs QDA

	DeLong's test for two correlated ROC curves

data:  model_rocs[[i]] and model_rocs[[j]]
Z = -8.4099, p-value < 2.2e-16
alternative hypothesis: true difference in AUC is not equal to 0
95 percent confidence interval:
 -0.3899051 -0.2425158
sample estimates:
AUC of roc1 AUC of roc2 
  0.3457371   0.6619475 
 

NB vs KNN

	DeLong's test for two correlated ROC curves

data:  model_rocs[[i]] and model_rocs[[j]]
Z = 6.5023, p-value = 7.908e-11
alternative hypothesis: true difference in AUC is not equal to 0
95 percent confidence interval:
 0.1639870 0.3055026
sample estimates:
AUC of roc1 AUC of roc2 
  0.7032221   0.4684773 
 

NB vs XGB

	DeLong's test for two correlated ROC curves

data:  model_rocs[[i]] and model_rocs[[j]]
Z = 9.058, p-value < 2.2e-16
alternative hypothesis: true difference in AUC is not equal to 0
95 percent confidence interval:
 0.1592489 0.2471954
sample estimates:
AUC of roc1 AUC of roc2 
  0.7032221   0.5000000 
 

NB vs ANN

	DeLong's test for two correlated ROC curves

data:  model_rocs[[i]] and model_rocs[[j]]
Z = -1.3725, p-value = 0.1699
alternative hypothesis: true difference in AUC is not equal to 0
95 percent confidence interval:
 -0.06293271  0.01109353
sample estimates:
AUC of roc1 AUC of roc2 
  0.7032221   0.7291417 
 

NB vs LDA

	DeLong's test for two correlated ROC curves

data:  model_rocs[[i]] and model_rocs[[j]]
Z = -1.121, p-value = 0.2623
alternative hypothesis: true difference in AUC is not equal to 0
95 percent confidence interval:
 -0.05415297  0.01474607
sample estimates:
AUC of roc1 AUC of roc2 
  0.7032221   0.7229256 
 

NB vs QDA

	DeLong's test for two correlated ROC curves

data:  model_rocs[[i]] and model_rocs[[j]]
Z = 1.9802, p-value = 0.04769
alternative hypothesis: true difference in AUC is not equal to 0
95 percent confidence interval:
 0.0004210301 0.0821281572
sample estimates:
AUC of roc1 AUC of roc2 
  0.7032221   0.6619475 
 

KNN vs XGB

	DeLong's test for two correlated ROC curves

data:  model_rocs[[i]] and model_rocs[[j]]
Z = -1.202, p-value = 0.2294
alternative hypothesis: true difference in AUC is not equal to 0
95 percent confidence interval:
 -0.08292195  0.01987661
sample estimates:
AUC of roc1 AUC of roc2 
  0.4684773   0.5000000 
 

KNN vs ANN

	DeLong's test for two correlated ROC curves

data:  model_rocs[[i]] and model_rocs[[j]]
Z = -7.3365, p-value = 2.193e-13
alternative hypothesis: true difference in AUC is not equal to 0
95 percent confidence interval:
 -0.3303019 -0.1910269
sample estimates:
AUC of roc1 AUC of roc2 
  0.4684773   0.7291417 
 

KNN vs LDA

	DeLong's test for two correlated ROC curves

data:  model_rocs[[i]] and model_rocs[[j]]
Z = -7.0228, p-value = 2.175e-12
alternative hypothesis: true difference in AUC is not equal to 0
95 percent confidence interval:
 -0.3254616 -0.1834349
sample estimates:
AUC of roc1 AUC of roc2 
  0.4684773   0.7229256 
 

KNN vs QDA

	DeLong's test for two correlated ROC curves

data:  model_rocs[[i]] and model_rocs[[j]]
Z = -5.332, p-value = 9.714e-08
alternative hypothesis: true difference in AUC is not equal to 0
95 percent confidence interval:
 -0.2645870 -0.1223534
sample estimates:
AUC of roc1 AUC of roc2 
  0.4684773   0.6619475 
 

XGB vs ANN

	DeLong's test for two correlated ROC curves

data:  model_rocs[[i]] and model_rocs[[j]]
Z = -10.531, p-value < 2.2e-16
alternative hypothesis: true difference in AUC is not equal to 0
95 percent confidence interval:
 -0.2717891 -0.1864943
sample estimates:
AUC of roc1 AUC of roc2 
  0.5000000   0.7291417 
 

XGB vs LDA

	DeLong's test for two correlated ROC curves

data:  model_rocs[[i]] and model_rocs[[j]]
Z = -9.9116, p-value < 2.2e-16
alternative hypothesis: true difference in AUC is not equal to 0
95 percent confidence interval:
 -0.2670080 -0.1788431
sample estimates:
AUC of roc1 AUC of roc2 
  0.5000000   0.7229256 
 

XGB vs QDA

	DeLong's test for two correlated ROC curves

data:  model_rocs[[i]] and model_rocs[[j]]
Z = -6.8895, p-value = 5.599e-12
alternative hypothesis: true difference in AUC is not equal to 0
95 percent confidence interval:
 -0.2080193 -0.1158758
sample estimates:
AUC of roc1 AUC of roc2 
  0.5000000   0.6619475 
 

ANN vs LDA

	DeLong's test for two correlated ROC curves

data:  model_rocs[[i]] and model_rocs[[j]]
Z = 0.62961, p-value = 0.5289
alternative hypothesis: true difference in AUC is not equal to 0
95 percent confidence interval:
 -0.01313458  0.02556686
sample estimates:
AUC of roc1 AUC of roc2 
  0.7291417   0.7229256 
 

ANN vs QDA

	DeLong's test for two correlated ROC curves

data:  model_rocs[[i]] and model_rocs[[j]]
Z = 3.6661, p-value = 0.0002462
alternative hypothesis: true difference in AUC is not equal to 0
95 percent confidence interval:
 0.03127138 0.10311699
sample estimates:
AUC of roc1 AUC of roc2 
  0.7291417   0.6619475 
 

LDA vs QDA

	DeLong's test for two correlated ROC curves

data:  model_rocs[[i]] and model_rocs[[j]]
Z = 3.6091, p-value = 0.0003072
alternative hypothesis: true difference in AUC is not equal to 0
95 percent confidence interval:
 0.02786361 0.09409248
sample estimates:
AUC of roc1 AUC of roc2 
  0.7229256   0.6619475 
 
