Back to all posts

2013: The Road Ahead at Quantopian

2013 is here! We've got a lot planned for the coming year and wanted to share some highlights of what you can expect to see from us over the next several months.

At Quantopian we are committed to providing quants the tools and support needed to learn, create, and reap the benefits of quantitative finance, all as easily accessible as your browser. That is why we are building the world’s first algorithmic trading platform in your browser. It’s a big vision, and we are laser-focused on building this out in 2013.

Algorithmic trading can be broken down into three broad steps: discovery, testing, and trading. The first step is where ideas are born. It’s where quants, armed with volumes of data begin a static analysis – slicing and dicing information in a variety of ways, trying different connections and searching for interesting patterns. Once an idea has crystallized, the next step is to express it as an algorithm. It’s at this point that the idea is put through its paces using real market data for backtesting. While these tests are underway the idea and the algorithm can be refined and optimized. When the creator is confident with how their idea will perform, it’s time for it to be put to the ultimate test, with real dollars on the line through a brokerage account.

At Quantopian we started by focusing on that middle step – the phase where algorithm development and testing take place. We've built tools to make it easy for people to get started writing algorithms, and have combined those tools with a backtester to let them use up-to-date data to test performance. We've been thrilled with the response and are continuing to improve the backtester, while we build support for the next step: live trading.

Early in 2013 we’ll be adding new paper trading capabilities to allow everyone to test their algorithms against live up-to-the-minute pricing data. This “walk forward testing” will let you hook up your algorithm and see how it performs in the market without executing any actual trades. We’ll also be making some graphing improvements to include prices and signals on our backtest graphs for better visualization of your algorithm.

Here are a few other exciting developments coming in 2013:

Backtester Updates: Our open-source backtester, Zipline, is being updated regularly. Almost every week we put out a new release with bug fixes, expansions, etc., and in 2013 we’ll be building on this progress with even more updates. The libraries will be expanded to include things such as ta-lib and even more risk metrics (Sharpe ratio, Sortino ratio, and so on.) And of course we’ll continue to work on improving the performance of the backtester to run your backtests even faster.

Education: One of the things we’re really excited about introducing in 2013 is the launch of our education area. We will provide resources for people just getting into algorithmic trading to learn more about quantitative finance; help people who don’t know much about Python to learn how to get started coding; and give experienced users tips and tricks from recognized experts. The site will have a series of structured tutorial lessons and will be packed with useful insights and information. It will be a work in progress as we continue to add materials, but expect to see the first lessons arriving later this winter.

Security: Finally, in 2013 we will continue to invest heavily in security. Among other things, we will be evaluating LXC and other sandbox-style security measures to determine the most effective solution, and will be encrypting all user algorithms. As we get closer to providing live trading capabilities it will become even more important to protect the intellectual property of our community. It’s already our top priority, and we’ve heard from many of you just how critical it is. Providing the level of security needed isn’t a static task that can be achieved and then forgotten; we’ll be working constantly to provide the safest and most secure development environment we can provide.

Well down the road, our plans for the future of Quantopian go beyond live trading of US stocks. Our data sources will expand to include options, futures, Forex and so on, until the full range of trading possibilities are supported.  We are excited about providing the data, but we're focused on live trading first.

Hopefully you’re feeling as excited about 2013 as we are. 2012 provided a fantastic starting point for Quantopian; we expect that the developments we have planned – and the new levels of engagement they will support – are going to make the coming year an incredibly exciting one.



The material on this website is provided for informational purposes only and does not constitute an offer to sell, a solicitation to buy, or a recommendation or endorsement for any security or strategy, nor does it constitute an offer to provide investment advisory or other services by Quantopian. In addition, the content of the website offers no opinion with respect to the suitability of any security or any specific investment.

Quantopian makes no guarantees as to accuracy or completeness of the views expressed in the website. The views are subject to change, and may have become unreliable for various reasons, including changes in market conditions or economic circumstances. All investments involve risk – including loss of principal. You should consult with an investment professional before making any investment decisions.

Michael

Do you have any plans to handle passive order placement in zipline?

By passive order placement, do you mean limit orders? Limit orders are on the roadmap for sure.

If you meant something else, let me know!

Michael

I meant limit orders. I think it would be nice to have limit orders and a probability of getting filled when a trade price hits your limit order and completely filled when the quote moves into your order. The easy example would be most market making strategies would take over the earth if you were passive and got every trade in which the trade price lifts your passive order so it would be nice to be able to adjust the parameter.

It would be interesting for shorter term stat arb models to understand what level of fills I would need in order to make the system work.

It also would be really nice to have an idea when you plan to be able to be used for live trading either via FIX or via some broker platform.

Limit orders are on our roadmap, for sure. They are on the same path as live trading of US equities - we're working very hard towards that. I'm not ready to share a timeframe yet, unfortunately. We just haven't done enough estimation work for me to have a date I believe in.

In the long run we should be able to amass quite a bit of fill data - what orders fill in what circumstances, and which ones don't. Before we put that into the product we're going to have to make sure we can anonymize the data enough to protect what our members' privacy. I think that is quite a ways away. It's an exciting one to think about, thought.

Comments are closed.