install.packages("quantmod")
library("quantmod")

getSymbols("AMZN", src = "yahoo", from = "2019-01-01", to = "2025-05-30")
getSymbols("AAPL", src = "yahoo", from = "2019-01-01", to = "2025-05-30")
getSymbols("GOOG", src = "yahoo", from = "2019-01-01", to = "2025-05-30")
getSymbols("MSFT", src = "yahoo", from = "2019-01-01", to = "2025-05-30")
getSymbols("META", src = "yahoo", from = "2019-01-01", to = "2025-05-30")
getSymbols("TSLA", src = "yahoo", from = "2019-01-01", to = "2025-05-30")
getSymbols("NVDA", src = "yahoo", from = "2019-01-01", to = "2025-05-30")

price = cbind(AMZN$AMZN.Adjusted,AAPL$AAPL.Adjusted,GOOG$GOOG.Adjusted,
              MSFT$MSFT.Adjusted,META$META.Adjusted,TSLA$TSLA.Adjusted,
              NVDA$NVDA.Adjusted)
ret   = apply(log(price),2,diff)
p     = ncol(ret)
n     = nrow(ret)


price1 = scale(price)

par(mfrow=c(1,1))
ts.plot(price1,col=1:p)
legend("topleft",legend=names,col=1:p,bty="n",lty=1,cex=0.75)

ret = apply(log(price),2,diff)

par(mfrow=c(2,4))
for (i in 1:p)
  ts.plot(ret[,i],ylab="",main=names[i],ylim=range(ret))


# Principal components analysis
# -----------------------------
pca = princomp(ret)

summary(pca)

par(mfrow=c(2,4))
for (i in 1:p)
  plot(pca$scores[,1],ret[,i],xlab="1st principal component",ylab=names[i])


# Factor analysis
# ---------------
fa = factanal(ret,factors=1,scores="regression")

fa

par(mfrow=c(1,1))
ts.plot(fa$scores[,1],ylab="Common factor")

par(mfrow=c(1,1))
plot(fa$scores[,1],pca$scores[,1],xlab="1st principal component",ylab="Common factor")

cbind(fa$load,fa$uniq)