Back to Community
Where do I go from here? What to do with risk model output?

First off, thanks Q for the awesome new risk model! I've been running old algorithms through it, especially ones that I thought had potential, looking to see if there really was something to them. Attached is the risk model for one such algo.

This result gives me hope for the signal. There are lots of specific returns with a decent sharpe, but they are eroded by common factors. Now all I have to do is figure out a way to reduce my exposure to these factors, but I'm not really sure where to start. My gut reaction is to load up all the common factors in the pipeline, set some contraints, and use optimize to build 0 exposure portfolios, but that seems a bit naive.

Any suggestions would be greatly appreciated!

Loading notebook preview...
1 response

Glad you like it! Overall, we're working on adding more education and tools to help you with these set of questions. You're asking the right questions.

One of the hard things about algo writing is that there is no straightforward "do these 3 steps and you're done" from here. What I'd suggest you do first is think about your alpha factor and these results. Sometimes one can come up with an alpha factor, it looks great, but then when you do performance attribution you find out you simply rediscovered the momentum factor. When that happens you just have to throw it out and start over. It looks like you have high exposure to a couple of styles - do you understand why? Do you think your factor is intrinsically linked to them? Or do you think you can remove those common factors and still hang on to your alpha?

For instance, it looks like you're pretty long the size factor. Can bin your universe into some high-cap, mid-cap, and small-cap stocks and then apply the alpha property to each bin, thereby smoothing out your size bias? Or is your alpha directly linked to the size of the stocks?

Then, you should do pretty much what you're thinking of. Add the risk model to your optimizer and see if you can constrain it away. For some algos that will work, for others it will constrain away the alpha, too, and then you're stuck. Note - we're working at adding API improvements so you can add the risk model directly into your simulation.


The material on this website is provided for informational purposes only and does not constitute an offer to sell, a solicitation to buy, or a recommendation or endorsement for any security or strategy, nor does it constitute an offer to provide investment advisory services by Quantopian. In addition, the material offers no opinion with respect to the suitability of any security or specific investment. No information contained herein should be regarded as a suggestion to engage in or refrain from any investment-related course of action as none of Quantopian nor any of its affiliates is undertaking to provide investment advice, act as an adviser to any plan or entity subject to the Employee Retirement Income Security Act of 1974, as amended, individual retirement account or individual retirement annuity, or give advice in a fiduciary capacity with respect to the materials presented herein. If you are an individual retirement or other investor, contact your financial advisor or other fiduciary unrelated to Quantopian about whether any given investment idea, strategy, product or service described herein may be appropriate for your circumstances. All investments involve risk, including loss of principal. Quantopian makes no guarantees as to the accuracy or completeness of the views expressed in the website. The views are subject to change, and may have become unreliable for various reasons, including changes in market conditions or economic circumstances.