Quantopian's community platform is shutting down. Please read this post for more information and download your code.
Back to Community
Expanding the universe of instruments + future road map + general retail punter quant trading in the real world

Greetings,

I am UK based, and work in the pointy end of finance. On the subject of what instruments to expand to next, I wanted to make my suggestion for Futures. you have the depth of market/liquidity, massive choice of contracts and I believe they massively lend themselves to quant/trend. Then obviously London listed equities :p

On the subject of being UK based, I am able to sign up to IB (its just moving £10k out of my normal broker to IB) but I just wondered if at some point you guys were planning on putting a high level road map as to where you want to take this project. I am deeply committed to making a venture into quant finance my next major project, but want to make sure I am backing the best horse I can (if that makes sense). Learning a language is no small feat and I want to stay relevant.

What I want : To learn a relatively future proof language, that has the right tools and community, open source and appropriate connectivity to the real world and brokers! While I am not changing my career, this is my passion and has been since my degree in Comp Sci.

What I have played with so far:

AmiBroker, NinjaTrader, MetaTrader, R, TSSB software. All have their merits and seem to be really popular, but just don't have the X factor. Some just crunch data (TSSB), some just arent as accessible as Python/Quantopian, some I just think they feel a little retail and not professional/enough potential.

Anybody else at the same juncture as me? Or can add anything?

Sorry, sunday morning brain dump. Happy Easter.

2 responses

Hello tt,

Have you looked at cAlgo yet? I don't really have any C# knowledge so I haven't got very far yet.

http://www.spotware.com/products/client-side-applications/calgo

P.
(Failed sea-side town, UK)

I was going to look into MATLAB, but with some elbow grease, I was able to get a python stack to do a lot of very cool stuff. Getting iPython qtconsole going was a major leap forwards for interactive data analysis, with zipline and picloud for parallel optimizations.

For a production system, I think it would depend on the periodicity, for HFT, I don't think I would want to use anything other than C++ -- Ernie Chan recently mentioned LimeTrader, but I haven't looked into it. For a daily+ system, I would either want to automate it in something like Quantopian, or else a little more manually as this guy did here: http://sanzprophet.blogspot.com/2013/03/strategies-on-cloud-taa-on-google-docs.html