Hi I am looking for an open source software to develop/backtest/forward-test trading strategies. I need the software to be as open as possible so I can read data from different data sources and integrate to different trading platforms. Any suggestion is welcome!
Well, the hard part is not reading data or integrating with different trading platforms. The hard part is finding an edge that worths the time spent to learn the software.
If you don't want to answer his/her question, why post? OP: Look at open source projects like TradeLink and ActiveQuant for starters.
I answered indirectly but you are one of those that think others should answer the way they think it is proper to answer. I will answer the way I want to answer. If you don't like that, put me on ignore. I already put you smart as*.
Thanks! both of you are correct! Actually AQ is java and really confusing for me. After half an hour visiting the website I could not find out where is the core framework (AQ) upon which AQR and other tools are built! But TraderLink seems promising as I am a .NET developer. I will take a look!
Hi! I think tradelink is my choice! Although it has some limitations but the core engine is flawless and it gives enough flexibility for a strategy designer to implement his ideas.
Please report back with your experiences on TradeLink. I'm very interested in how the learning curve is with it. I'm considering migrating my IDE (OpenQuant) development there once I've gotten things tested and debugged (it seems like a similar event architecture)
I'm not exiting OpenQuant, I am finding it decent and plan to continue to use it. It does a nice job of bridging the gap from backtesting -> paper trading -> live trading. It's got enough structure placed around it in areas such as data collection, charting, performance measurement etc... I was a bit overwhelmed by the amount of flexibility (aka lack of structure) that TradeLink exposed, and I'm relatively novice at .NET development so I prefer something a bit more contained. This is not implying that TL would be complex for others with more experience. I like the idea of TradeLink being lighter footprint, and more flexible. There are some clumsy approaches required in OQ to trade multiple symbols in a portfolio (pairs trading) and I like the idea of being able to build a software stack that could, in the future, be independent of a 3rd party vendor. It seems like TL is very similar in the basic framework of events (GotTick instead of OnQuote) etc... so I'm hopeful that TL would be a good execution platform once I've gotten everything running the way I'd like Feel free to PM for more detail.