#Labb2 - Assignment 2
require(tree)
rm(list=ls())
library(readxl)
data <- read_excel("creditscoring.xls")
set.seed(12345)

n=dim(data)[1]
id=sample(1:n, floor(n*0.5))
train=data[id,]
test=data[-id,]

nv=dim(train)[1]
idv=sample(1:nv, floor(nv*0.5))
test=train[idv,]
valid=train[-idv,]

#2
model_dev = tree(as.factor(good_bad)~.,train,split = "deviance")
pred_model_dev = predict(model_dev,train,type="class")
summary(model_dev)

model_gin = tree(as.factor(good_bad)~., data =train,split = "gini")
pred_model_gin = predict(model_dev,train,type="class")
summary(model_gin)
#3
# cv.res=cv.tree(model_dev)
# plot(cv.res$size, cv.res$dev, type="b",
#      col="red")
# plot(log(cv.res$k), cv.res$dev,
#      type="b", col="red")
# min_gin=min(cv.res)

#optimal leaves for deviance-fit

fit=model_dev
trainScore=rep(0,9)
testScore=rep(0,9)
for(i in 2:9) {
  prunedTree=prune.tree(fit,best=i)
  pred=predict(prunedTree, newdata=valid,
               type="tree")
  trainScore[i]=deviance(prunedTree)
  testScore[i]=deviance(pred)
}
plot(2:9, trainScore[2:9], type="b", col="red",
     ylim=c(200,600))
points(2:9, testScore[2:9], type="b", col="blue")