Quantopian is home to 120,000 people learning algorithmic trading, including students, data scientists, academic researchers, developers, and finance professionals.
We provide a research platform, market simulation, and data for free. We also provide tutorials, community, and lectures to teach you how to get good at it. I recommend you take a look at the Getting Started Guide (https://www.quantopian.com/tutorials/getting-started) and then start going through the Lectures (https://www.quantopian.com/lectures). The lectures cover some important statistical topics, and they get into how to apply those concepts to algorithmic trading.
Quantopian's revenue model is to build a hedge fund and charge the fund investors returns/management fees. The algorithms in the hedge fund come from the Quantopian community. We work with the best algorithm writers on our platform, negotiate compensation, and then put their code to work.
We would be crazy to charge people to use our platform. We need thousands of algorithms, and charging for the platform would be one of the faster ways to kill our business.
Yes, you can code to Interactive Broker's API. But where would you get your free minute historical data for backtesting? Or corporate fundamentals data? Or the free IPython research environment? Or the community of 60,000 quants giving each other mentoring and advice?
I work at Quantopian, so you can imagine my answer to all of those questions.
There's a slightly different methodology here, but one consistent with what you're looking for.
On one line, buy-and-hold the S&P 500. Re-invest all dividends. You are 100% in the market at all times.
On the other line, buy-and-hold all companies run by female CEOs, weighted by the number of companies. Rebalance your portfolio every time a company is added or removed. You are 100% in the market at all times.
I think that if you look at the IPython notebook that the Fortune article refers to you can find the details spelled out in code.
(FWIW, the calculations are done by a she, not a he! It's my colleague Karen.)
We work very hard to make our interests aligned with our community members' interests. We don't literally make money from the algorithms in the contest. What we're doing is encouraging hundreds and thousands of new people to write algorithms. The best ones will be invited to join our hedge fund, we'll negotiate compensation with them, and we'll take investment from pension funds and endowments and the like. In that sense - yes, when our customers win, we win too.
As for the judging, I think you'll find it be very transparent. There are 6 return and risk metrics calculated for both the backtest and the paper trading. The 6 metrics are weighted equally to generate an overall score. You can see the metrics for every contestant and the combined score on the leaderboard. You can verify it all by downloading the CSV. https://www.quantopian.com/leaderboard
It's both forward-testing and backward testing. The algos have been locked since submission - some were submitted as early as 1/15, all were submitted by 2/2. That makes it both an in-sample and out-of-sample test.
Yes, the Quantopian platform includes default commissions. It also includes default slippage. No model is perfect, of course, but this is a tool that's had a lot of development.
I think you're looking at the trees, and you should step back and look at the forest.
The fraction of CEOs that are women is dramatically smaller than the fraction of the population that are women. There is no qualitative explanation as to why that should be true. So long as that remains true, it's worth looking into why it is true. The relative performance of the group is fair game for investigation.
You retain ownership of the content you put in our system; everything you write is yours. Your intellectual property remains private and your own. You can read more about our policies in our terms and in our FAQ (https://www.quantopian.com/faq and https://www.quantopian.com/policies/terms)
Of course, there is no way that we can prove or guarantee that we're not peeking. Like anything else in the cloud, at some point it becomes a matter of trust. That's why we did our About page a bit differently (https://www.quantopian.com/about). We're all startup veterans with reputations in the industry. You can click the links there and find out who we know in LinkedIn, and see what they say about us. We hope that our good reputations make it easier for you to trust us. Of course, that is entirely up to you!
Another solution is to write the algorithm but avoid the hedge fund - lots of suits, and they take most of the money. You're better off if you trade it yourself.
People work for hedge funds because hedge funds provide mentorship and really powerful tools and lots of data to sift through. We're trying to provide all of those things, for free, at Quantopian. https://www.quantopian.com/ Check out our community (for mentorship), our backtester (very powerful, and open source), and our 11-years of minute-bar data - all for free.
Very neat story. Thanks for writing it up. The nuts and bolts of this kind of operation are fascinating. I know a lot of people out there have trading ideas, but don't really know how to implement them.
If you just finished the post and you want to try coding up a trading algorithm for free, check out http://www.quantopian.com. It uses Python, not MQL, and provides free data and backtesting. (Yes, I work at Quantopian)
Aneth. . . . you're reading our mind. Come back to Quantopian on Thursday. I think you'll like what you see.
Yes, there are people who trade today and make money using algorithms. They are few and far between, mostly because the toolset is so hard to build. Data, backtester, trading platform, etc. all take a long time to build. We're trying to make it much easier by providing all the tools. You need an idea; we'll make the rest work for you.
The benchmark on Quantopian is indeed modeled after the S&P 500.
How would the algo do in 2008? It's trivial for you to check it yourself. Click the "clone algo" button, change the time range of the test, and click "Run Backtest." Question answered!
Cash management is built already. We track how much you have, dividend payments, all that stuff. We've built many risk measurements, too: alpha, beta, Sortino, Information Ratio, etc.
Risk management is far more complex. Risk management is more a part of the algorithm itself than a feature that we can build. That said, we can add more risk tools. We're very open to suggestions, if you have some in mind.
Yeah. This algo is highly leveraged - like 15X. It's possible to really lose your shirt if you trade this algo exactly.
Taibo's algo is interesting as a starting point. It's not one that that you just take off the shelf and start trading with. But, you can take it and learn from it and develop an alternative strategy. Presumably one with less risk!
There aren't a ton of hurdles left before we start offering "live trading" on Quantopian. We have all the pieces, we just need to stitch them together. A couple more months, I think.
In the beginning, at least, it will be leveraged through your existing brokerage account. You're going to integrate Quantopian with your brokerage, and Quantopian will place orders for you with your brokerage.
If we're as successful as we hope to be that will mean we're driving a lot of trading volume. If you start driving enough trading volume, the exchanges start to pay you rather than the other way around. It would be a pretty sweet day if we can offer trading for free to our members and fund the company on the exchange fees.
The the thing about backtesting a strategy is that it is very easy to make a mistake in your backtester. Look ahead bias is the most common mistake.
Another challenge is the data. Are you testing against a history of stocks that includes bankruptcies? If not you have survivorship bias.
I suggest you take a look at my website, www.quantopian.com. Look at our open-sourced backtester, https://github.com/quantopian/zipline. Between the two we can help you get past those two sources of error.
We provide a research platform, market simulation, and data for free. We also provide tutorials, community, and lectures to teach you how to get good at it. I recommend you take a look at the Getting Started Guide (https://www.quantopian.com/tutorials/getting-started) and then start going through the Lectures (https://www.quantopian.com/lectures). The lectures cover some important statistical topics, and they get into how to apply those concepts to algorithmic trading.
disclosure: I work for Quantopian.