See also the list of quantitative trading firms.
You will become a quant by getting a job as a quant.
I know that’s kind of redundant and not particularly useful, but it’s the only approach.
You can read books to get some terminology, but don’t get too bogged down on this. What really matters are your technical skills. It especially helps if you can demonstrate your skills publicly, like with your thesis or an open-source project.
Just apply for roles immediately. It takes a long time to land that first job out of school, so start the process now.
I get this question often and I don’t like it because there isn’t much to learn from it. It’s like asking, “How did you meet your spouse?” as if there is a specific or repeatable path someone could take.
But I’ve written about my personal experiences before, so here it is again:
I first found-out about quantitative finance during my PhD program. Like most grad students, I was looking for an excuse to procrastinate from my dissertation. I began looking at the course catalog of different departments until I saw a course entitled "The Mathematics of Finance".
The syllabus listed terminology I had never seen before, like Black Scholes. When I read-up on what options pricing was, I realized that this was an excellent application of my academic specialty: high-performance computing.
I applied to every job ad at a finance company that required technical credentials. I eventually got a position on a bank’s prop desk.
Again, this story probably won’t help you. Just concentrate on your technical skills, especially if you can show them off publicly.
I advise people against getting too wedded to a specific tech stack, asset class, or time horizon early in their career. Despite learning of this industry through Black Scholes, I have never traded options professionally.
Every firm has their own boundaries for what defines a role. And there can be a lot of overlap of responsibilities. That said, here is one possible way of looking at some categories:
-
Analyst: researchers who work on models, optimize parameters, and investigate post-trade activity
-
Developer: engineers who write production-quality software; might focus on a piece of the pipeline like data, analytics, or execution
-
Trader: people who make decisions on parameters and handle the unexpected
Some firms blend research and engineering into a “strategist” role, while others strictly segregate those positions.
At some firms the trader’s duty is merely to operate the system, where at others the trader is the manager of the team.
There is tremendous variety to what a “quant” might do.
The best way to see the problems that professional quants run into is to look at the Quantitative Finance Stack Exchange.
(Caveat: That is a website for people already in the industry. If you are not in the industry yet, do not post on there. Do not ask on that website how to become a quant; you will be ejected.)
That said, you can still read the site and see what issues people run into. Sometimes the challenges are about programming, sometimes they’re about strategy, sometimes they’re about math.
So feel free to read a few posts to get an idea for what is interesting to you. And please do not post on there if you are not already working in the industry.
This is heavily dependent on the firm. A lot of my career has been in high-frequency trading. The best analogy I have for that is motorsports. HFT is a game of infrastructure, just like F1: the most important aspect is the car, not the driver.
On the other hand, there are plenty of strategies that are not infrastructure heavy and involve making decisions while sitting in front of a spreadsheet. Here the driver is the only thing that matters.
Either way, I still recommend getting good at programming because you will be able to automate away a lot of tedious tasks and perform interesting research and optimizations.
Again, this is heavily dependent on the firm. The most common work involves just getting data into a usable format, which is still programming. When it comes to number crunching, there's time-series analysis, Black Scholes, portfolio optimization, and machine learning. Which techniques you use will depend on asset class and time horizon.
Only do a PhD if you really want to learn about that topic.
Grad school is years of living at the poverty line to build something nobody wants. It's a labor of love.
If you want to go into finance, just develop your technical skills and then join a trading firm. There is no need to do a degree in an unrelated field from the job that you actually want.
Again, this is personal to me and your mileage may vary, but pursuing a PhD expanded my technical skills and got me a lot of practice with scholarly writing. I did it because I wanted to learn more about high-performance computing.
But I didn't know about quantitative finance before I started grad school; it wasn't until halfway through that I learned about this career possibility. Then I had to slog through a thesis that I no longer cared about, which was extremely taxing for both me and my supervisors.
If you want to become a quant, then go do that. If you want to learn about a topic within a graduate program, then pursue that too. But don't do a degree that is unrelated to the field you would rather be in. That doesn't help anybody.
The other common credential beyond a PhD is the CFA.
The CFA covers a lot of material, particularly in the fundamental space. However, the CFA does not cover exotic options or high-frequency trading. So this credential's desirability depends heavily on what the firm is doing.
Quantitative finance is really broad. HFT and exotics have no overlap, so puting together a list of what-to-learn-to-be-a-quant doesn’t work. It's like creating a master knowledge base for all of programming without narrowing the field to web, embedded, gaming, etc.
The only way to learn is to get a job in the field. Hone your technical skills first. You can read a finance book to learn some terminology (it doesn’t matter which one), but everything you’ll need to know about the industry will come from learning on the job. It’s far more important to demonstrate your math and programming skills upfront.
Whatever your boss pays you to use.
Every language you can think of is used by at least one trading firm. C++ and Python are very common, but there are also firms that use Haskell, OCaml, q/kdb+, R, Java, etc.