library(tseries)
library(quantmod)
library(forecast)
library(car)
library(haven)
 
index3<-getSymbols("^TWII",auto.assign = FALSE)
index.open<-na.omit(data.frame(index3[,1]))
index.close<-na.omit(data.frame(index3[,4]))
index.open.test<-data.frame(index.open[1:(nrow(index.open)-365),])
index.close.test<-data.frame(index.close[1:(nrow(index.open)-365),])
index.open.train<-data.frame(index.open[1:(nrow(index.open)),])
index.close.train<-data.frame(index.close[1:(nrow(index.open)),])
index.open.year<-data.frame(tail(index.open,365))
index.close.year<-data.frame(tail(index.close,365))
colnames(index.open.year)="OP.value"
colnames(index.open.train)="OP.value"
colnames(index.close.year)="close.value"
colnames(index.close.train)="close.value"
mod1<-auto.arima(index.open.train, seasonal = TRUE,ic="aic",test = "adf",seasonal.test ="seas",allowdrift = TRUE,
                 allowmean = TRUE,stepwise=FALSE,approximation=FALSE)
mod2<-auto.arima(index.close.train, seasonal = TRUE,ic="aic",test = "adf",seasonal.test ="seas",allowdrift = TRUE,
                 allowmean = TRUE,stepwise=FALSE,approximation=FALSE)
predict.open<-forecast(index.open.test,model=mod1,h=365,include.mean = TRUE)
predict.close<-forecast(index.close.test,model=mod2,h=365,include.mean = TRUE)
d2=0
for(x in c(1:365)){d1<-(index.open.year[x]-index.close.year[x])-(predict.close$fitted[x]-predict.open$fitted[x])
d2<-sum(d1)}
print(d2)

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