Quantopian's community platform is shutting down. Please read this post for more information and download your code.
Back to Community
First day.... WHERE TO START? Help

So I have through a variety of sources become very interested in quantitative models. I've consumed as many articles and books as possible in the last several months, however I am struggling to find a place to really dig in and start.

What is the best method to begin making and experimenting with my own algorithm? Any tips in general as I embark on the journey into the quant world?

1 response

Hi Will -

It depends on what you'd like to accomplish. If you are interested in the Q contest/fund, the framework to have in mind is described here:

https://blog.quantopian.com/a-professional-quant-equity-workflow/

Q has built up tools in all of the critical areas of the workflow, and they tend to be supportive if you can't find something or need help using a tool.

Basically, if you can construct such an algorithmic beast that meets the criteria described on https://www.quantopian.com/allocation, you'll have as good a shot as anyone in getting an allocation (however, patience is required, since as I understand, Q is very firm about seeing at least 6 months of out-of-sample backtest performance).

On a micro-level, Q recently has made it clear that they'd like to see alpha factors incorporating data sets that are non-traditional (basically not OHLCV bar and fundamental data, as I understand their request). I suspect that it is well-known within the industry that technical indicators will only get you so far in constructing a long-short equity fund. One thing to keep in mind is that even though some of the Q data sets don't extend to the current date (unless you pay), my understanding is that in 6 months, Q will re-run any backtest and evaluate it, using the full dataset. So, even though you can't enter such algos into the contest, you can still get them evaluated behind the scenes for the Q fund.

On the other hand, if you are interested in trading your own money (and I have no experience in this area), it is a matter of how much capital you intend to apply, and how to define your risk-reward mix, and the rationale for it.