The future of quantitative trading

Posted: March 20, 2010 in .NET, C#, GPUs

Programming languages:

The industry until now has been dominated by C (lesser extent), C# (the current market), and C++ ( the old King and close runner up).  I believe I am seeing a gradual shift in the current marketplace though.  It seems like many friends of mine are jumping ship and going to OCaml or its knock off brother, F#.  I must admit that I recently downloaded F# ( from Microsoft Research and liked it.

Like it or now, I believe that F# is going to take over the functional programming world (I am sadden too here; Haskell/Erlang guy right here!) and with the .NET integration with other products already this isn’t unreasonable considering greater than 50% of the trading shops are running windows .

Modeling software:

This has been all over the place in the last few years from R, Matlab, SAS, Mathematica, …… but I feel that the industry has chopped it down to a close 3: R, Matlab, and Python (use I called it under Modeling software….I know it is a scripting language but quants love it and programmers hate it just like LISP).  Most hedge funds and prop shops seem to go with Matlab while amateur traders and academia prefers R.  I actually prefer R because it is free and has many of the same add-ins that the Matlab toolboxes provide if you are willing to dig some for them.

I think we will see a larger migration to R as the packages become more robust and we see companies ease back on expenditures.

Trading platforms:

I used to see many traders perfectly content with packages like Xtrader and CQG for their trading but it seems like over the last 5 years much of the industry has went to their own in-house platform solutions as the need for advanced trading strategy integration and execution speed has become a concern.  While I have personally used many customizable off the shelf advanced platforms like RTS and Orc, it appears even they are losing the battle to in-house  software investments.  I personally really like Orc’s software too!

There are many open source code options for developers to go off of but my personal recommendation for anyone trying to get in the field is to come out of the college with a decent knowledge of C# and be prepared to learn OCaml/F# in a few years.  Also focus on market micro structure as that is the realm where these trading algorithms live for the most part; other quants typically handle the large picture with co-integration, covariance, distribution estimation models, etc.

I believe we will see Trading Technologies, CQG, and others lose much of their commercial industry support as the bulk goes in-house.  I am not sure how they are going to combat it easily but I see it coming based on the type of work that I have done/or been involved in recent years.

  1. I thought it made sense to reach out to you and introduce you to Lightspeed Trading. Headquartered in New York City, Lightspeed Trading is a leading provider of ultra low latency DMA trading technology, risk management solutions, and brokerage services for professional retail active traders, proprietary trading groups, hedge funds, and institutional investors. Lightspeed Trading offers both a software-based solution, Lightspeed Trader, for clients employing electronic-based trading strategies and an automated trading solution, Lightspeed Gateway, for clients employing algorithmic or “black box” trading strategies.

    We’d love to talk with you more about becoming a listing partner on your blog and for any articles/info for your blog. In addition, if you ever write about Lightspeed or have your readers ask questions, please feel free to contact me for information.


  2. […] The future of quantitative trading March 2010 2 comments 3 […]

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s