I think Ruby needs to start broadening its Appeal beyond Rails.
Introduction to Zipline: A Trading Library for Python
151–155 of 155 posts
Re: Introduction to Zipline: A Trading Library for Python
#152Earlier quoted context omitted.
Who's going to provide that data for free when they can charge a LOT of money for access to it. Ever notice why free data on Google or Yahoo are 15 min delayed?
I know what the exchanges charge for data access, and even if you pay it is in an antiquated format. Tradier has been providing equity and option tick data for free for years in a clean RESTful API with the capability of websockets and streaming. So thats the answer to your question of who. Like I said, in this town we call it disruption.
I am a developer that has done a few projects on the tradier API. Agreed
Re: Introduction to Zipline: A Trading Library for Python
#153Earlier quoted context omitted.
Wrong that's all you need. In fact according to Peter Lynch you have a better chance of alpha as you dont have to deal with all the bs a PM at a big fund has too.
The markets that Peter Lynch traded and the markets today are completely different.
Sigh, trading vs. investing... Also if you make a statement like that I am sure you are unfamiliar with his thesis and methodology.
Re: Introduction to Zipline: A Trading Library for Python
#154Earlier quoted context omitted.
2 questions - what kind of leg work is involved, and how much is a fair bit of money? Is it possible with 250k of working capital?
Leg work: - Getting the data into a shape that you can use. Normally a total PITA. For futures, you have to either stitch the contracts yourself, or get a pre-stitched series, which you have to take time to understand. Filtering it for weird data points. - Writing the strategy / backtesting code. The fun part. - Connecting to a broker. Gotta read API docs, test the functions, connect it to your code in a way that mak…
If you're talking about HFT and front running (which HFTs do whether you like it or not) a retail trader can't compete in that sector.
HFT operates on algorithms that mostly involve making money on the spread by running ahead of the brokers, buying the cheap stuff, and selling it to the broker who needs it. They don't trade on market microstructure, mostly because all of it starts to fall apart at the tick level.
250k is CERTAINLY enough to invest in futures contracts. You can do it with much less. Much much less. Futures are highly leveraged instruments. You can diversify by trading multiples of futures contracts (or e-minis depending on account size) because you're only required to post initial and maintenance margin.
You can easily blow up your account, but if you're just TRADING 5000 is enough to start selling a few contracts. If you're looking to start building something sustainable 10k is enough. But, the more the better.
Re: Introduction to Zipline: A Trading Library for Python
#155General question: how do you take this and interact directly with the market? Is there some sort of general, public api that you're making calls agains, where do you get an account for it, etc? Or, is this going through some firm that interfaces with the market?
Upvotes, but no comments :(
If anyone comes back to this ever, still interested in knowing!