Getting Started with Open Source for Quantitative Finance

Connect with Jacques on LinkedIn

During my senior year, at a university that promised to teach me things, I found myself staring at financial statements wondering to myself “what is it the Quants see that I don’t?” and so began my journey into quantitative finance.

Throughout my re-education I have come across several members of the QF community that provide leadership and structured learning. Two of them are The Python Quants and the CQF Institute which gave me the opportunity to attend their recent “For Python Quants Conference” in New York, via live streaming.

Dr. Yves Hilpisch’s talk on Open Source in QF caught my attention; it was the first lecture and the one I chose to capture in writing. I wrote this article after attending the For Python Quants conference and the ideas written here are not my own but Dr. Hilpisch’s, I did add a few extra comments where I could. I have also attached a link to the slideshow he used during the presentation: Click Here.

Mega Trends

So then let’s get started. When Dr. Hilpisch started his talk he opened with the mega trends that are influencing QF and then split up into 3 main topics of interest which I will cover:

  1. The social aspect
  2. The technological aspect
  3. The educational aspect


1. The Social aspect: Meet Up groups and other topics


Online communities are gaining popularity as the global community becomes smaller. He mentioned as a niche questions and answers forum for professionals and academics alike. Just looking at the website now I see it riddled with members interacting over topics like: pricing options, backtesting methods, forecasting, and risk adjusted volatility.


Another great source is Meet Up groups, which is all about what people do offline. It is a place where like minded individuals can come together to share ideas and network, in a real – actual place, not online. Dr Hilpisch has taken the initiative by starting a group called Python for Quant Finance, for those of you looking to get started.

Then there are of course the conferences that as the demand increases. Examples of this are Quantcon and ForPythonQuants.


The one thing he didn’t mention but that I’d like to add is the large blogging community around QF, the topics range from building an event-driven backtester to using machine learning as a way to optimise a portfolio. The three main blog aggregators are Quantocracy, QuantNews, and my website QuantsPortal.


2. The Technological aspectLanguages

Next he covered how open source languages are dominating the field of data science.

The commercial languages are losing ground as more and more institutions are moving to open source languages. Please note: this doesn’t mean that the financial industry isn’t still a secretive place but rather that institutions are making use of the vast options of free financial analytic libraries out there.


In 2014 Python was the most popular coding language and so it becomes a strategic decision to consider using it going forward. He pointed out that due to Python being very easy to learn many US universities are using it in their courses as an introductory language. As a South African citizen I can tell you the majority of my friends, who studied computer sciences, are using Python today.

You will also find that service providers are using Python as a marketing tool to communicate quantitative concepts to clients. An example of this is Eurex that now provides their clients with a variance advanced services tutorial, which is about variance futures that they released last year September. The main reason they chose Python was because they couldn’t expect their clients to pay for commercial software to trade with them as well as how easy it is to learn.

One of the downsides to Python is that it doesn’t have a very strong array of financial libraries, with a few exceptions like the Zipline package from Quantopian. This is probably attributed to financial institutions not wanting to share trade secrets.

If you are looking for packages, he did recommended searching Github. One of the gems he recently found is TSTables which allows you to store high frequency data in a very systematic fashion.


Financial libraries in other languages

Julia Logo

Julia: Although Julia has a very small community, it is a very active community that has already built many useful things for quantitative finance. Dr. Hilpisch recommends taking a look.


Lua: is another upcoming language for QF which is light weight and boasts high performance. It is one of the fastest just-in-time compilers around.

R Logo

R: probably has the most to offer when it comes to open source and financial analytic capabilities. There are many – many libraries available in the R world which is probably the reason why it is so popular amongst data scientists.


3. The Educational aspect

Speaking from experience here I can tell you that I agree with Dr. Hilpisch when he says that there are no real books or courses around open source for Quantitative Finance. When I got started I looked for the CFA equivalent and only recently did the University of Johannesburg launch an honours programme for students to bridge into QF.

There are a few single dedicated books on the topic including:

  • “Python for Finance”(O’Reilly)
  • “Derivatives Analytics with Python” (Wiley Finance)
  • I would also recommend “Algorithmic Trading and DMA”

For those readers looking for formal education in the field:

Final Remarks

A great big shout out to Dr. Hilpisch: thank you for sharing your experiences with us. If you would like to contact Dr. Yves Hilpisch, here are his details:

For those of you interested in reading more on quantitative finance and machine learning, I encourage you to check out my blog and content aggregator

Feel free to leave a comment below. I would be very interested to find the other ways in which beginners are getting started.


0 replies

Leave a Reply

Want to join the discussion?
Feel free to contribute!

Leave a Reply

Your email address will not be published. Required fields are marked *