Well if you intend to find a job as quant researcher you will need a PhD or at least a master's in a quantitative subject.

If you want to go independent then don't do it. It takes years of study and you might not find anything substantial.

You're correct (all quant roles do require a masters in a quantitative subject in their JD), but I have an MS (Physics) and a past 2-month internship in quant research, so there's no question that I'm going to do it. I just wanted some guidance on how to prepare in the best possible way.

I tend to disagree a bit with the degree requirement. In general if you demonstrate impressive talent/research, firms will be interested. Firms that are stubborn and judge you just based on your on-paper degrees are likely firms you don't want to work for anyway. I would focus incredibly strongly on building your rigorous statistical knowledge, both in the specific tests, but more importantly in the general way a statistician looks at the world. Statisticians are like scientists and tend to constantly be looking for reasons to doubt claims: Why is this potentially not profitable? What biases could be causing this to falsely look good? Even Jim Simons of RenTech says that quant is really mainly statistics (https://www.youtube.com/watch?v=QNznD9hMEh0).

In order to rapidly iterate on statistical research you'll need to learn more programming. Both Python and R are very useful, but Python has a bit more momentum right now in industry and in my experience is generally far easier to use than R. Plus you can use Python for more applications, like full stack development or front end if you work in tech.

I want to shamelessly pitch our lecture series as it covers a lot of stats and coding. Folks I've talked to have said that having research done on Quantopian has helped them get jobs, as it's a way to display a portfolio just like an artist might. Quant firms will want to see interesting pieces of research more so than completed algorithms usually.

At the end of the day pick projects that are interesting to you. Google what you don't know and keep pushing forward. The thing is, if you develop some great skills in stats and programming, there are tons of jobs waiting to be filled right now, not just quant. So it's not an over-concentrated investment to learn these things.

Thanks a lot for the detailed answer! I am indeed going through the various videos on the Quantopian YT channel (mainly the talks) and intend to get to the lecture series after I've seen the Quantopian videos that are more general (not specific to Quantopian platform). I feel that I'll be in a better position to use the platform that way.

As for Statistics, could you please suggest any good books to build up a rigorous statistical knowledge? As of now, I'm planning on reading Probability Theory: The Logic of Science by Jaynes. Any other reference recommendations apart from that? I'm asking this because I can't discount the possibility that there may be some important material not covered in the lecture series or videos.

Finally (and maybe this question is the same as above), could you recommend a comprehensive source from where I can read up on the specific tests used in algo trading model evaluation, if any such source exists?

Our lectures are actually designed to be agnostic of the platform and teach statistical and quantitative concepts, so ideally you wouldn't have that problem. If you find any lecture to be too platform specific please let me know, the curriculum is used by professors who want to teach the concepts independent of platform, so we want to be sure that the platform isn't taking center stage.

I'm not a book learner, I learned most of what I know by search around online, reading wikipedia articles, chatting with folks, etc. I can't recommend any books as a result. Also, one of the reasons we designed the lecture series is I didn't know of any good resource for the stats necessary for quantitative finance. A lot of the other resources are slides from various universities or even original academic papers. I'll let my colleague Max answer with a few books he likes.

I think if I was your age I would turn my skills to something more intrinsically satisfying and useful than quantitative finance. 😀🤗😆😇

@Delaney : That would be very nice of you, thanks!

Hey Shirish,

A few of my favorite books on stats are:

• "Statistical Inference", by Casella and Berger
  
• "Financial Time Series Analysis", by Tsay
  
• "The Elements of Statistical Learning", by Hastie (et al.)
  
• "Bayesian Methods for Hackers", by Davidson-Pilon (which you can actually find on github here) (I also helped out with the PyMC3 version of this)

Those are all pretty solid books, with focuses on various different aspects of statistics. I hope that you find them useful!

For Statistics and Econometrics, I really like Econometrics: Modern Approach by Wooldridge. The book has nice appendices that summarize important concepts from Statistics, Math and linear algebra. For Time series econometrics, I could not really find a good textbook though but the lecture series here at Quantopian is very rich with regards to Time Series.