#LDA
library(readxl)
## Warning: package 'readxl' was built under R version 3.6.1
data <- read_excel("C:/Users/Administrator/Downloads/Data Bank.xlsx") #importdata
View(data)
summary(data)
## COLLATERAL CHARACTER CAPACITY CONDITION
## Min. :1.250 Min. :-8.248 Min. :1.000 Min. :1.65
## 1st Qu.:2.500 1st Qu.: 4.668 1st Qu.:2.976 1st Qu.:1.65
## Median :3.750 Median : 5.201 Median :3.277 Median :1.65
## Mean :3.537 Mean : 4.974 Mean :3.278 Mean :1.77
## 3rd Qu.:3.750 3rd Qu.: 5.676 3rd Qu.:3.583 3rd Qu.:1.65
## Max. :7.000 Max. : 6.400 Max. :4.786 Max. :5.36
## CAPITAL kode KOL
## Min. : 4.000 Min. :0.0000
## 1st Qu.: 5.000 1st Qu.:0.0000
## Median : 8.000 Median :1.0000
## Mean : 8.745 Mean :0.6147
## 3rd Qu.:10.000 3rd Qu.:1.0000
## Max. :20.000 Max. :1.0000
data = as.data.frame(data)
n = dim(data)[1]
Y = as.matrix(data[,6])
#discriminantanalysis
library(lda)
library(MASS)
## Warning: package 'MASS' was built under R version 3.6.2
library(caret)
## Warning: package 'caret' was built under R version 3.6.2
## Loading required package: lattice
## Loading required package: ggplot2
## Warning: package 'ggplot2' was built under R version 3.6.2
model=lda( data[,6] ~ ., data[,1:5])
model
## Call:
## lda(data[, 6] ~ ., data = data[, 1:5])
##
## Prior probabilities of groups:
## 0 1
## 0.3852814 0.6147186
##
## Group means:
## COLLATERAL CHARACTER CAPACITY CONDITION CAPITAL
## 0 4.148989 4.740983 3.141124 1.931292 8.353933
## 1 3.153944 5.120715 3.364276 1.668680 8.989437
##
## Coefficients of linear discriminants:
## LD1
## COLLATERAL -0.660192460
## CHARACTER 0.029602165
## CAPACITY 1.285572537
## CONDITION -0.961248155
## CAPITAL -0.003063417
fit <- lda(data[,6] ~ ., data[,1:5], CV=TRUE)
compare = confusionMatrix(table(data[,6],fit$class))
compare
## Confusion Matrix and Statistics
##
##
## 0 1
## 0 190 166
## 1 40 528
##
## Accuracy : 0.7771
## 95% CI : (0.7488, 0.8035)
## No Information Rate : 0.7511
## P-Value [Acc > NIR] : 0.0357
##
## Kappa : 0.496
##
## Mcnemar's Test P-Value : <2e-16
##
## Sensitivity : 0.8261
## Specificity : 0.7608
## Pos Pred Value : 0.5337
## Neg Pred Value : 0.9296
## Prevalence : 0.2489
## Detection Rate : 0.2056
## Detection Prevalence : 0.3853
## Balanced Accuracy : 0.7934
##
## 'Positive' Class : 0
##
#CART
library(rpart)
library(rpart.plot)
## Warning: package 'rpart.plot' was built under R version 3.6.3
model1 = rpart(data[,6] ~ ., data[,1:5], method = "class")
# Plot the trees
rpart.plot(model1)
fit = predict(model1, data[,1:5], type = 'class')
compare2 = confusionMatrix(table(data[,6],fit))
compare2
## Confusion Matrix and Statistics
##
## fit
## 0 1
## 0 246 110
## 1 22 546
##
## Accuracy : 0.8571
## 95% CI : (0.8329, 0.8791)
## No Information Rate : 0.71
## P-Value [Acc > NIR] : < 2.2e-16
##
## Kappa : 0.6838
##
## Mcnemar's Test P-Value : 3.665e-14
##
## Sensitivity : 0.9179
## Specificity : 0.8323
## Pos Pred Value : 0.6910
## Neg Pred Value : 0.9613
## Prevalence : 0.2900
## Detection Rate : 0.2662
## Detection Prevalence : 0.3853
## Balanced Accuracy : 0.8751
##
## 'Positive' Class : 0
##