if (!require("pacman")) install.packages("pacman") pacman::p_load(timeSeries,fPortfolio,quantmod,caTools,Quandl,PerformanceAnalytics,fAssets) library(timeSeries); library(fPortfolio); library(quantmod); library(caTools);library(PerformanceAnalytics)# may also
"OPTIMIZE YOUR PORTFOLIO" library("devtools") library(quantmod) library(Quandl)
#load in the russel 2000 stocks setwd("C:/Users/jroac_000/Desktop/START-UP") russell <- read.csv("russell_2000_.csv") rus2000<- as.vector(russell$TICKER) #put into a vector so we can request
rus2000 <- rus2000[-906] #the reason I have do this individually is because there are tickers in the string that the API can't pull, thus stoppping the process ClosingPricesRead <- NULL for (Ticker in rus2000) ClosingPricesRead <- cbind(ClosingPricesRead, getSymbols(Ticker, src = "google",from = "2017-06-01")) ClosingPrices <- ClosingPricesRead[apply(ClosingPricesRead,1,function(x) all(!is.na(x))),] returns <- as.timeSeries(tail(ClosingPrices,-1) / as.numeric(head(ClosingPrices,-1)) - 1)
install.packages("corrplot") #ALL_CORR<- all_correlations(returns,sorted="strength",type="spearman") library(corrplot) #TEMP <- data.frame( t(as.matrix(RESULTS2[,2:12])) ) #names(TEMP) <- sectors correlations <- abs(cor(returns)) row_indic <- apply(correlations, 1, function(x) sum(x > 0.3 | x < -0.3) > 1) correlations<- correlations[row_indic ,row_indic ] corrplot(correlations[,-13], method="square", title = "Industrial Sector") top_rank_industrial_corr <- as.data.frame(sort(apply(correlations,1,mean)))
#BACK TEST getSymbols("SPY", src = "google",from = "2017-01-01")