Quantopian – A Quant Trader’s Best Friend, Connors Research Traders Journal (Volume 55):
Christopher Cain, CQF, CMT
U.S. Quantitative Equity Strategist at Bloomberg LP
How would you like to build trading strategies, do research, improve your existing strategies, plus get access to free historical data including fundamental data?
How To Improve Your Trading with Quantopian
Quantopian is a leading website to learn quantitative finance, practice your Python programming skills, do high-level quantitative research, backtest trading algorithms and do a deep analysis of your historical test results.
In this issue of the Connors Research Traders Journal, we are going to introduce you to the Quantopian platform and touch on some of the many benefits this platform provides. This is not designed to be a complete overview of what Quantopian has to offer, but rather an introduction to the platform and to highlight some of its benefits.
We believe Quantopian, and the powerful tools it provides, can help make you a better trader.
What is Quantopian?
At its heart, Quantopian is a place to conduct quantitative research and build trading strategies.
Quantopian provides access to high-quality pricing data for stocks, ETFs and futures and doesn’t require the user to download the many Python packages that are needed to perform this kind of analysis.
As long as you do your research/strategy development on the Quantopian website, there are no downloads or other technical hurdles for you to overcome.
In addition to pricing data, Quantopian also allows access to fundamental data provided by Morningstar. You can use this fundamental data to construct trading strategies!
Two Environments
There are two main environments in which to work in Quantopian – the notebook environment and the Interactive Development Environment (IDE), where you actually test your trading strategies.
The notebook environment is an embedded Jupyter notebook, which has quickly become the ubiquitous tool for quants everywhere.
In this environment, you can grab historical data from Quantopian’s database and look for historical edges, run statistical or machine learning models, make various plots/visuals and much more.
The main point of the notebook environment is to do initial research into your trading idea before constructing a full-blown trading strategy.
The second environment is the IDE, where you actually code your trading logic.
Here you do things such as pull in any security your algorithm plans to trade (including setting up a dynamically changing universe of securities), write logic that controls when to buy and sell, as well as write code that controls portfolio level logic such as how many positions to have, sizing of the positions, etc.
Using the Pipeline
A rather unique thing about Quantopian is its ability to use “pipelines” in trading algorithms. Pipelines are a way to trade a dynamically changing universe of stocks, avoiding survivorship bias and other pitfalls that come with portfolio level testing.
The pipeline allows a user to calculate multiple statistics about all stocks in a universe and can be used to filter out the stocks you want your algorithm to trade.
You can use some of Quantopian’s built-in universes, such as the “Q500US” universe, which dynamically selected the 500 most liquid US stocks (determined by a trailing average dollar volume statistic) every month.
In 2019, testing strategies on one security at a time is not enough. You must test strategies on a portfolio level, using many securities. This is how professional quants and systematic traders develop their strategies.
Quantopian allows you to conduct this type of research.
Enter Fundamental Data
Another advancement Quantopian made is the availability and use of high-quality fundamental data provided by Morningstar. This data can be pulled at the security level, using a dynamic universe of securities such as the Q500US mentioned earlier.
For example, say you wanted to grab the 100 stocks with the lowest price/book ratio, of our 500 stock universe, every week. After getting the basics down, this is easily accomplished using Quantopian and Python, in just a few lines of code.
This opens up a whole new toolset for many traders, who haven’t had access to this data before. Even if a trader did have access to this data, the technical challenge of setting up a dynamic universe of stocks, calculating different statistics for hundreds of stocks, then filtering the ones you want, is likely beyond your technological skills.
Quantopian allows you to do this type of backtesting.
Combining Technical Analysis with Quantitative and Fundamental Analysis.
A few weeks ago we showed you a strategy we created that combined quantitative analysis with fundamental analysis. The test results were stunning and you can read that issue of the Connors Research Traders Journal here.
Detailed Portfolio Backtest Results
Another benefit of Quantopian is the ability to do a deep dive into the analysis for your backtested results. The Python package that Quantopian uses to analyze your backtest is called Pyfolio.
Pyfolio allows the user to write Python code to analyze your backtested results in any way you want, including custom metrics and graphs.
Pyfolio also comes with many preprogrammed statistics and “tear-sheets”, to allow the user to quickly produce a report of the historical test results of a given strategy.
The Benefits of Quantopian
We find Quantopian to be a wonderful platform to do quantitative research, build high powered trading strategies, and learn python coding specifically as it relates to quantitative finance and trading.
Just a handful of the many benefits Quantopian offers includes:
- “Batteries included” – there is no need to download anything when working on the Quantopian website.
- Free Data. This includes pricing data for US stocks and ETFs, fundamental data from Morningstar, and continuous futures data.
- Two environments – one to test the validity of an idea and one to take that idea and build a full-blown trading strategy around it. This makes your work more efficient.
- Ability to apply your trading logic on a dynamically changing, survivorship bias-free dataset of hundreds and possibly thousands of stocks at once.
- Ability to access fundamental data and use this data to filter out stocks you want to trade, opening a whole new potential avenue for traders to explore.
- Ability to do a deep dive into your historical test results.
- The ability to join a global community of quantitative traders, many of them whom work for asset management firms and professional trading firms.
This touches on a few of the many benefits Quantopian offers systematic traders. All of this is done in a modern environment, using Python, the most popular coding language used by professional quants and systematic traders.
We encourage you to give Quantopian a try – we believe it will ultimately make you a better, more successful trader.
If you’d like to learn how to program in Python on Quantopian’s platform in under 6 hours, we’re teaching a 5-week course beginning on September 10. Please see the details below.
More Knowledge To Improve Your Trading
There Is No Future For Traders Who Don’t Know Python – efinancialcareers.com
The majority of the professional trading desks around the world are requiring their traders to create and test their strategies using Python (not Amibroker and TradeStation). They’re doing this because Python allows their traders to build better strategies faster, and more efficiently, with the objective of increasing profits.
Please listen to this webinar and join the largest and fastest-growing community in the investment world who now test and program their strategies in Python. This webinar discusses how and why Python will improve your trading.
The Alpha Formula – High Powered Strategies to Beat The Market With Less Risk teaches you strategies and portfolios with historical Alpha in Stocks, ETFs, and Fixed Income.
Backed by dozens of academic studies and combining behavioral finance with Ray Dalio’s correlation research, this book will teach you new, easy to understand quant strategies you can apply immediately.