Hi, I'm beginning to fool around in R for trading strategy purposes. To keep it simple, I have only played with stock data at this point.
I have created a simple trend following strategy (in blue). Given my statistical background, I am attempting to bootstrap the results and table the same parameters highlighted below with no luck (in green). Any ideas on what I could do differently? Really Appreciate your help!!! Thanks library(quantmod) library(PerformanceAnalytics) b <- get(getSymbols('SPY'))["2011::"] s <- get(getSymbols('GLD'))["2011::"] b$sma1 <- SMA(Cl(s) , 1) s$sma50 <- SMA(Cl(s) , 50) s$position <- ifelse(Cl(s) > s$sma50 , 1 , -1) myReturn <- lag(s$position) * dailyReturn(s) table.Drawdowns(s$position, top = 5, digits = 1) table.Stats(s$position, ci = 0.95, digits = 2) table.SpecificRisk(s$position, b$sma1, Rf = 0, digits = 2) table.Correlation(s$position, b$sma1) charts.PerformanceSummary(cbind(dailyReturn(s),myReturn)) N = 100 # Number of simulations Loop = mat.or.vec(N,2,1,1,1) for (i in 1:N){ # sample with replacement from return distribution of index s.new = (sample(s, length(s), replace = T, prob = NULL)) # demeaning returns s.new = s.new-mean(s) # new price series starting at same value as original series prices.new = xts(prices[[1]]*exp(cumsum(s.new))) # define strategies # mean reversion s$sma50.new = SMA(Cl(s.new) , 50) # Create buy/sell signals # mean reversion s$position.new <- ifelse(Cl(s) > s$sma50.new , 1 , -1) # replace missing values with zeros s$position.new[is.na(s$position.new)] = 0 Loop[i,1] = if (mean(s$position.new) > mean(s$sma50.new)) {1}else{0} } #Loop # plots simulated series returns.new = cbind(s$sma50.new, cumsum(s$sma50.new)) chart.CumReturns(returns.new,s$sma50.new,geometric=F) Erik Schiele Vice President Money Markets Trading, Originations and Sales 101 Barclay St, NY NY 10007 3rd Floor BNY Mellon Capital Markets, LLC Main Desk 212-815-8222 This is for informational purposes only; from sources the Firm believes reliable; may not be accurate or complete; is subject to change; is not a recommendation or offer to buy/sell a financial instrument or adopt any investment strategy; is not legal, tax, credit or accounting advice. Do not use e-mail to submit any instructions - acceptances are at your risk. The Firm or its affiliates lends to, borrows from and provides other products/services to issuers and others, receives compensation therefore, and periodically has a direct or indirect financial interest in the financial instruments/transactions indicated. Additional risks may exist that are not referenced. Past performance is not indicative of future returns. Other than CDs or CDARS, financial instruments: are not FDIC insured; are not deposits or other obligations of and are not guaranteed by the Firm or any bank or non-bank affiliate; and involve investment risk including possible loss of principal. The Firm is ! a wholly owned, indirect non-bank subsidiary of The Bank of New York Mellon Corporation, and a member of FINRA and SIPC, and is solely responsible for its obligations and commitments. The information contained in this e-mail, and any attachment, is confidential and is intended solely for the use of the intended recipient. Access, copying or re-use of the e-mail or any attachment, or any information contained therein, by any other person is not authorized. If you are not the intended recipient please return the e-mail to the sender and delete it from your computer. Although we attempt to sweep e-mail and attachments for viruses, we do not guarantee that either are virus-free and accept no liability for any damage sustained as a result of viruses. Please refer to http://disclaimer.bnymellon.com/eu.htm for certain disclosures relating to European legal entities. [[alternative HTML version deleted]] ______________________________________________ R-help@r-project.org mailing list -- To UNSUBSCRIBE and more, see https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.