On Friday, 27 November 2015 at 23:26:27 UTC, karabuta wrote:
This question came into mind when I read this http://www.makeuseof.com/answers/which-programming-language-is-used-to-build-a-financial-trading-platform/
Quant is a big domain these days, and I don't think there is one answer to that question - it depends specifically on what you want to do, what resources you have to hand, and what your constraints are.
Andy Smith used D successfully at one of the largest and most respected hedge funds there is. See his dconf 2015 talk. And I myself am using D (helped by a couple of people from the community) for analytics within the financial area.
When they are ready they will be used to manage a decent pot of capital. I have the advantage of not having a large existing legacy code base or many other systems that I will need to hook into. There has been a bit of a price to pay to wrap libraries and APIs that I needed, but as Andy Smith says - it's actually strangely pleasurable when you get in the groove (for a C API), and in the beginning one learns more about the language by doing so. It's also a price you pay once and upfront (until they break the API in the next release, at least), and so it was something I could bear given I was the customer as well as the programmer. There are many larger corporate environments where this kind of delay or extra work wouldn't be acceptable, because people want results and now, and the short-term focus doesn't lend itself on picking the best tool for the longer-term. So people in that situation must consider their constraints carefully when making a decision.
The CTO of one decent sized hedge fund told me that a few years ago it would have been tough to accept using a language that wasn't widely used. He said today increasingly the theme from other CTOs he speaks to is that there is more pragmatism and a receptivity to different choices. Languages then become adopted internally when one guy has some success with something new and becomes an evangelist for it.
