Quant development: Who is using R to develop and test trading algorithms?
I am currently using AlgoTrader, Quantmod and TTR to develop and test trading strategies.
Quantmod provides some basic stuff but for real evaluation of a trading system, I need to push data tick by tick into the algorithm. AlgoTrader package is the best available that I have seen. This helps you develop and evaluate Algorithms in an object-oriented fashion, but lacks multi instrument support, position sizing and more complex risk management.
What other packages and/or approaches would you recommend to research and evaluate a robust system using R?
—– Have a look at quantstrat, blotter, FinancialInstrument and PerformanceAnalytics…
Also RBloomberg and IBrokers
i have used R to do some basic coding ..you can find details @ http://quantfinanceindia.blogspot.com/
Hi all, thanks for the replies. So far I’m impressed with R as a rapid prototyping tool. I am currently using Quantmod for charting (no zoom feature sadly), AlgoTrader for backtesting (good but no longer in development, I have the source though) and XTS for timeseries. I should add PerformanceAnalytics and FinancialInstrument to get the input more realistic & see the output.
I am interested in speeding the simulations hence might apply Rcpp to implement AlgoTrader packages inner loop in C.
Finally I’m interested in adding genetic optimisation around the overall strategy. Is there any predefined package that allows genetic optimisation of input parameters given rules of what constitutes “good” output parameters?
Try blotter’s packages (blotter, quantstrat etc.) + DEoptim for optimisation.
NOTE I now post my TRADING ALERTS into my personal FACEBOOK ACCOUNT and TWITTER. Don't worry as I don't post stupid cat videos or what I eat!