Cloud9Trader makes writing automated trading algorithms simple and accessible to those with little programming knowledge. It streamlines the workflow and makes the process of testing algorithms exponentially more efficient. It then runs your algorithms live, sending signals to your broker for trade execution.
Traders have historically used desktop software to develop and test their algorithms. Moving the workflow online offers you all the benefits of cloud computing. All your scripts are saved securely and versioned. All your backtest results and trades are stored for quick retrieval, your complete history. You won’t need to make backups - your data is safe and always available. You won’t need to leave your laptop running or set up expensive virtual machines and you will never miss a trade through lost connection whilst trading live.
Backtesting is the process of running your algorithm against real historical price data to determine if it is likely to be profitable. This is very processor intensive as the script must run for every price tick, typically millions over a normal backtest period. You won’t need to lock up your home computer while your processor chews through the data. Our platform allows you to run as many backtests simultaneously in the cloud as there are server instances available. You are free to continue working while your backtests are running. Additionally, our architecture fires up new servers automatically whenever there is demand, scaling back down when it is quiet.
Historical price data is unmanageably large. You would soon fill your hard drive if you were to deliver full tick data to backtests running locally. Existing software solutions circumvent this by only supplying price data at minute resolution. All the missing ticks are then just estimated. You rely completely on your backtest results when deciding whether to trade your algorithm live, so any inaccuracy is bad news.
We supply all the actual ticks, every single price update, as served by your broker for 100% like for like accuracy in backtesting. You can run at minute resolution to quickly check that your scripts are behaving as expected while you are developing them, though for complete accuracy you need to backtest against every tick. Algorithms often trade on very small margins and it was always a core tenet of ours that your backtesting should give you the exact result had you actually traded for that period. On completion, the results screen allows you to analyse your backtest run in every detail with rich visual feedback.
Optimization is the process of running backtests repeatedly to determine the set of external parameters that give the most profitable outcome when fed into your algorithm. Cloud9Trader includes a genetic optimization framework that models natural selection and genetic processes to direct the runner towards profitable results from a search space of potentially infinite possibilities. Each step is allocated its own cloud server so that your backtests run simultaneously, providing results in record time. The results let you see at a glance which are the most profitable and have the strongest equity curves. A good strategy externalizes the parameters that make it flexible for various instruments and market conditions. You should run these regularly to ensure your algorithm is at its peak configuration.
Take a look at Writing Your Technical Indicators for an introduction.
Our algorithm sharing infrastructure lets you share your algorithm code with the community. This allows you to pool resources and develop strategies collectively. If you need advice on your strategy or help with your code you can make it available for the community for their support and feedback. Algorithms you share are posted on our community forum as a base for discussion. Of course any algorithms you don’t share are kept private and owned by you.
If you are new to automated trading, this also means there is plenty of code available for you to use as a starting point for building your own strategies. It is often simplest to learn by example.
You can see what is shared at this moment on the Community Shared Algorithms page.
A quick intro on the general structure of Cloud9Trader algorithms.
A one page guide to writing your own custom technical indicators.
Was this page useful? If you find any errors or have any questions please get in touch at firstname.lastname@example.org.