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

RF vs LR

	DeLong's test for two correlated ROC curves

data:  model_rocs[[i]] and model_rocs[[j]]
Z = -1.5998, p-value = 0.1096
alternative hypothesis: true difference in AUC is not equal to 0
95 percent confidence interval:
 -0.051979900  0.005259056
sample estimates:
AUC of roc1 AUC of roc2 
  0.7055389   0.7256630 
 

RF vs SVM

	DeLong's test for two correlated ROC curves

data:  model_rocs[[i]] and model_rocs[[j]]
Z = 2.1407, p-value = 0.0323
alternative hypothesis: true difference in AUC is not equal to 0
95 percent confidence interval:
 0.003153861 0.071568124
sample estimates:
AUC of roc1 AUC of roc2 
  0.7055389   0.6649415 
 

RF vs DT

	DeLong's test for two correlated ROC curves

data:  model_rocs[[i]] and model_rocs[[j]]
Z = 2.4105, p-value = 0.01593
alternative hypothesis: true difference in AUC is not equal to 0
95 percent confidence interval:
 0.009451853 0.091688723
sample estimates:
AUC of roc1 AUC of roc2 
  0.7055389   0.6517322 
 

RF vs NB

	DeLong's test for two correlated ROC curves

data:  model_rocs[[i]] and model_rocs[[j]]
Z = -0.23348, p-value = 0.8154
alternative hypothesis: true difference in AUC is not equal to 0
95 percent confidence interval:
 -0.03703468  0.02915045
sample estimates:
AUC of roc1 AUC of roc2 
  0.7055389   0.7062447 
 

RF vs KNN

	DeLong's test for two correlated ROC curves

data:  model_rocs[[i]] and model_rocs[[j]]
Z = 1.7904, p-value = 0.07339
alternative hypothesis: true difference in AUC is not equal to 0
95 percent confidence interval:
 -0.003193514  0.070630070
sample estimates:
AUC of roc1 AUC of roc2 
  0.7055389   0.6685843 
 

RF vs XGB

	DeLong's test for two correlated ROC curves

data:  model_rocs[[i]] and model_rocs[[j]]
Z = 0.04075, p-value = 0.9675
alternative hypothesis: true difference in AUC is not equal to 0
95 percent confidence interval:
 -0.01913705  0.01994971
sample estimates:
AUC of roc1 AUC of roc2 
  0.7055389   0.7018962 
 

RF vs ANN

	DeLong's test for two correlated ROC curves

data:  model_rocs[[i]] and model_rocs[[j]]
Z = -0.57177, p-value = 0.5675
alternative hypothesis: true difference in AUC is not equal to 0
95 percent confidence interval:
 -0.04573713  0.02507845
sample estimates:
AUC of roc1 AUC of roc2 
  0.7055389   0.7126319 
 

RF vs LDA

	DeLong's test for two correlated ROC curves

data:  model_rocs[[i]] and model_rocs[[j]]
Z = -1.379, p-value = 0.1679
alternative hypothesis: true difference in AUC is not equal to 0
95 percent confidence interval:
 -0.04993537  0.00868929
sample estimates:
AUC of roc1 AUC of roc2 
  0.7055389   0.7229256 
 

RF vs QDA

	DeLong's test for two correlated ROC curves

data:  model_rocs[[i]] and model_rocs[[j]]
Z = 2.0471, p-value = 0.04065
alternative hypothesis: true difference in AUC is not equal to 0
95 percent confidence interval:
 0.001718287 0.078991721
sample estimates:
AUC of roc1 AUC of roc2 
  0.7055389   0.6619475 
 

LR vs SVM

	DeLong's test for two correlated ROC curves

data:  model_rocs[[i]] and model_rocs[[j]]
Z = 3.6209, p-value = 0.0002935
alternative hypothesis: true difference in AUC is not equal to 0
95 percent confidence interval:
 0.02785369 0.09358914
sample estimates:
AUC of roc1 AUC of roc2 
  0.7256630   0.6649415 
 

LR vs DT

	DeLong's test for two correlated ROC curves

data:  model_rocs[[i]] and model_rocs[[j]]
Z = 3.3581, p-value = 0.0007848
alternative hypothesis: true difference in AUC is not equal to 0
95 percent confidence interval:
 0.03078078 0.11708064
sample estimates:
AUC of roc1 AUC of roc2 
  0.7256630   0.6517322 
 

LR vs NB

	DeLong's test for two correlated ROC curves

data:  model_rocs[[i]] and model_rocs[[j]]
Z = 1.5048, p-value = 0.1324
alternative hypothesis: true difference in AUC is not equal to 0
95 percent confidence interval:
 -0.005872763  0.044709375
sample estimates:
AUC of roc1 AUC of roc2 
  0.7256630   0.7062447 
 

LR vs KNN

	DeLong's test for two correlated ROC curves

data:  model_rocs[[i]] and model_rocs[[j]]
Z = 2.657, p-value = 0.007884
alternative hypothesis: true difference in AUC is not equal to 0
95 percent confidence interval:
 0.01497397 0.09918343
sample estimates:
AUC of roc1 AUC of roc2 
  0.7256630   0.6685843 
 

LR vs XGB

	DeLong's test for two correlated ROC curves

data:  model_rocs[[i]] and model_rocs[[j]]
Z = 1.6657, p-value = 0.09577
alternative hypothesis: true difference in AUC is not equal to 0
95 percent confidence interval:
 -0.004198473  0.051731977
sample estimates:
AUC of roc1 AUC of roc2 
  0.7256630   0.7018962 
 

LR vs ANN

	DeLong's test for two correlated ROC curves

data:  model_rocs[[i]] and model_rocs[[j]]
Z = 0.81211, p-value = 0.4167
alternative hypothesis: true difference in AUC is not equal to 0
95 percent confidence interval:
 -0.01841853  0.04448070
sample estimates:
AUC of roc1 AUC of roc2 
  0.7256630   0.7126319 
 

LR vs LDA

	DeLong's test for two correlated ROC curves

data:  model_rocs[[i]] and model_rocs[[j]]
Z = 0.51576, p-value = 0.606
alternative hypothesis: true difference in AUC is not equal to 0
95 percent confidence interval:
 -0.00766501  0.01313977
sample estimates:
AUC of roc1 AUC of roc2 
  0.7256630   0.7229256 
 

LR vs QDA

	DeLong's test for two correlated ROC curves

data:  model_rocs[[i]] and model_rocs[[j]]
Z = 3.6491, p-value = 0.0002632
alternative hypothesis: true difference in AUC is not equal to 0
95 percent confidence interval:
 0.02949285 0.09793800
sample estimates:
AUC of roc1 AUC of roc2 
  0.7256630   0.6619475 
 

SVM vs DT

	DeLong's test for two correlated ROC curves

data:  model_rocs[[i]] and model_rocs[[j]]
Z = 0.5165, p-value = 0.6055
alternative hypothesis: true difference in AUC is not equal to 0
95 percent confidence interval:
 -0.03691561  0.06333420
sample estimates:
AUC of roc1 AUC of roc2 
  0.6649415   0.6517322 
 

SVM vs NB

	DeLong's test for two correlated ROC curves

data:  model_rocs[[i]] and model_rocs[[j]]
Z = -2.392, p-value = 0.01676
alternative hypothesis: true difference in AUC is not equal to 0
95 percent confidence interval:
 -0.075145834 -0.007460382
sample estimates:
AUC of roc1 AUC of roc2 
  0.6649415   0.7062447 
 

SVM vs KNN

	DeLong's test for two correlated ROC curves

data:  model_rocs[[i]] and model_rocs[[j]]
Z = -0.15932, p-value = 0.8734
alternative hypothesis: true difference in AUC is not equal to 0
95 percent confidence interval:
 -0.04845434  0.04116891
sample estimates:
AUC of roc1 AUC of roc2 
  0.6649415   0.6685843 
 

SVM vs XGB

	DeLong's test for two correlated ROC curves

data:  model_rocs[[i]] and model_rocs[[j]]
Z = -2.042, p-value = 0.04116
alternative hypothesis: true difference in AUC is not equal to 0
95 percent confidence interval:
 -0.072425320 -0.001484004
sample estimates:
AUC of roc1 AUC of roc2 
  0.6649415   0.7018962 
 

SVM vs ANN

	DeLong's test for two correlated ROC curves

data:  model_rocs[[i]] and model_rocs[[j]]
Z = -2.5012, p-value = 0.01238
alternative hypothesis: true difference in AUC is not equal to 0
95 percent confidence interval:
 -0.08506127 -0.01031939
sample estimates:
AUC of roc1 AUC of roc2 
  0.6649415   0.7126319 
 

SVM vs LDA

	DeLong's test for two correlated ROC curves

data:  model_rocs[[i]] and model_rocs[[j]]
Z = -3.6583, p-value = 0.0002539
alternative hypothesis: true difference in AUC is not equal to 0
95 percent confidence interval:
 -0.08904962 -0.02691845
sample estimates:
AUC of roc1 AUC of roc2 
  0.6649415   0.7229256 
 

SVM vs QDA

	DeLong's test for two correlated ROC curves

data:  model_rocs[[i]] and model_rocs[[j]]
Z = 0.17714, p-value = 0.8594
alternative hypothesis: true difference in AUC is not equal to 0
95 percent confidence interval:
 -0.03013269  0.03612072
sample estimates:
AUC of roc1 AUC of roc2 
  0.6649415   0.6619475 
 

DT vs NB

	DeLong's test for two correlated ROC curves

data:  model_rocs[[i]] and model_rocs[[j]]
Z = -2.3179, p-value = 0.02045
alternative hypothesis: true difference in AUC is not equal to 0
95 percent confidence interval:
 -0.100606614 -0.008418193
sample estimates:
AUC of roc1 AUC of roc2 
  0.6517322   0.7062447 
 

DT vs KNN

	DeLong's test for two correlated ROC curves

data:  model_rocs[[i]] and model_rocs[[j]]
Z = -0.69227, p-value = 0.4888
alternative hypothesis: true difference in AUC is not equal to 0
95 percent confidence interval:
 -0.06456366  0.03085964
sample estimates:
AUC of roc1 AUC of roc2 
  0.6517322   0.6685843 
 

DT vs XGB

	DeLong's test for two correlated ROC curves

data:  model_rocs[[i]] and model_rocs[[j]]
Z = -2.4845, p-value = 0.01297
alternative hypothesis: true difference in AUC is not equal to 0
95 percent confidence interval:
 -0.08973765 -0.01059026
sample estimates:
AUC of roc1 AUC of roc2 
  0.6517322   0.7018962 
 

DT vs ANN

	DeLong's test for two correlated ROC curves

data:  model_rocs[[i]] and model_rocs[[j]]
Z = -2.5741, p-value = 0.01005
alternative hypothesis: true difference in AUC is not equal to 0
95 percent confidence interval:
 -0.10727034 -0.01452892
sample estimates:
AUC of roc1 AUC of roc2 
  0.6517322   0.7126319 
 

DT vs LDA

	DeLong's test for two correlated ROC curves

data:  model_rocs[[i]] and model_rocs[[j]]
Z = -3.2518, p-value = 0.001147
alternative hypothesis: true difference in AUC is not equal to 0
95 percent confidence interval:
 -0.1141040 -0.0282827
sample estimates:
AUC of roc1 AUC of roc2 
  0.6517322   0.7229256 
 

DT vs QDA

	DeLong's test for two correlated ROC curves

data:  model_rocs[[i]] and model_rocs[[j]]
Z = -0.41553, p-value = 0.6778
alternative hypothesis: true difference in AUC is not equal to 0
95 percent confidence interval:
 -0.05839827  0.03796770
sample estimates:
AUC of roc1 AUC of roc2 
  0.6517322   0.6619475 
 

NB vs KNN

	DeLong's test for two correlated ROC curves

data:  model_rocs[[i]] and model_rocs[[j]]
Z = 1.6654, p-value = 0.09584
alternative hypothesis: true difference in AUC is not equal to 0
95 percent confidence interval:
 -0.006661674  0.081982461
sample estimates:
AUC of roc1 AUC of roc2 
  0.7062447   0.6685843 
 

NB vs XGB

	DeLong's test for two correlated ROC curves

data:  model_rocs[[i]] and model_rocs[[j]]
Z = 0.26513, p-value = 0.7909
alternative hypothesis: true difference in AUC is not equal to 0
95 percent confidence interval:
 -0.02779710  0.03649399
sample estimates:
AUC of roc1 AUC of roc2 
  0.7062447   0.7018962 
 

NB vs ANN

	DeLong's test for two correlated ROC curves

data:  model_rocs[[i]] and model_rocs[[j]]
Z = -0.3392, p-value = 0.7345
alternative hypothesis: true difference in AUC is not equal to 0
95 percent confidence interval:
 -0.04329388  0.03051943
sample estimates:
AUC of roc1 AUC of roc2 
  0.7062447   0.7126319 
 

NB vs LDA

	DeLong's test for two correlated ROC curves

data:  model_rocs[[i]] and model_rocs[[j]]
Z = -1.3121, p-value = 0.1895
alternative hypothesis: true difference in AUC is not equal to 0
95 percent confidence interval:
 -0.041598312  0.008236464
sample estimates:
AUC of roc1 AUC of roc2 
  0.7062447   0.7229256 
 

NB vs QDA

	DeLong's test for two correlated ROC curves

data:  model_rocs[[i]] and model_rocs[[j]]
Z = 2.5001, p-value = 0.01242
alternative hypothesis: true difference in AUC is not equal to 0
95 percent confidence interval:
 0.009569995 0.079024245
sample estimates:
AUC of roc1 AUC of roc2 
  0.7062447   0.6619475 
 

KNN vs XGB

	DeLong's test for two correlated ROC curves

data:  model_rocs[[i]] and model_rocs[[j]]
Z = -1.6365, p-value = 0.1017
alternative hypothesis: true difference in AUC is not equal to 0
95 percent confidence interval:
 -0.073208726  0.006584831
sample estimates:
AUC of roc1 AUC of roc2 
  0.6685843   0.7018962 
 

KNN vs ANN

	DeLong's test for two correlated ROC curves

data:  model_rocs[[i]] and model_rocs[[j]]
Z = -2.1287, p-value = 0.03328
alternative hypothesis: true difference in AUC is not equal to 0
95 percent confidence interval:
 -0.084603789 -0.003491449
sample estimates:
AUC of roc1 AUC of roc2 
  0.6685843   0.7126319 
 

KNN vs LDA

	DeLong's test for two correlated ROC curves

data:  model_rocs[[i]] and model_rocs[[j]]
Z = -2.4687, p-value = 0.01356
alternative hypothesis: true difference in AUC is not equal to 0
95 percent confidence interval:
 -0.09748404 -0.01119859
sample estimates:
AUC of roc1 AUC of roc2 
  0.6685843   0.7229256 
 

KNN vs QDA

	DeLong's test for two correlated ROC curves

data:  model_rocs[[i]] and model_rocs[[j]]
Z = 0.2996, p-value = 0.7645
alternative hypothesis: true difference in AUC is not equal to 0
95 percent confidence interval:
 -0.03677975  0.05005321
sample estimates:
AUC of roc1 AUC of roc2 
  0.6685843   0.6619475 
 

XGB vs ANN

	DeLong's test for two correlated ROC curves

data:  model_rocs[[i]] and model_rocs[[j]]
Z = -0.60547, p-value = 0.5449
alternative hypothesis: true difference in AUC is not equal to 0
95 percent confidence interval:
 -0.04548783  0.02401649
sample estimates:
AUC of roc1 AUC of roc2 
  0.7018962   0.7126319 
 

XGB vs LDA

	DeLong's test for two correlated ROC curves

data:  model_rocs[[i]] and model_rocs[[j]]
Z = -1.4295, p-value = 0.1529
alternative hypothesis: true difference in AUC is not equal to 0
95 percent confidence interval:
 -0.049863125  0.007804386
sample estimates:
AUC of roc1 AUC of roc2 
  0.7018962   0.7229256 
 

XGB vs QDA

	DeLong's test for two correlated ROC curves

data:  model_rocs[[i]] and model_rocs[[j]]
Z = 2.0315, p-value = 0.0422
alternative hypothesis: true difference in AUC is not equal to 0
95 percent confidence interval:
 0.001406807 0.078490542
sample estimates:
AUC of roc1 AUC of roc2 
  0.7018962   0.6619475 
 

ANN vs LDA

	DeLong's test for two correlated ROC curves

data:  model_rocs[[i]] and model_rocs[[j]]
Z = -0.67167, p-value = 0.5018
alternative hypothesis: true difference in AUC is not equal to 0
95 percent confidence interval:
 -0.04033139  0.01974399
sample estimates:
AUC of roc1 AUC of roc2 
  0.7126319   0.7229256 
 

ANN vs QDA

	DeLong's test for two correlated ROC curves

data:  model_rocs[[i]] and model_rocs[[j]]
Z = 2.7398, p-value = 0.006148
alternative hypothesis: true difference in AUC is not equal to 0
95 percent confidence interval:
 0.01442639 0.08694230
sample estimates:
AUC of roc1 AUC of roc2 
  0.7126319   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 
 
