Back to Community
Macro economic data for machine-learning models

I was wondering if anyone has an opinion on whether one can feed in macro-economic data, such as inflation, into ML models using the nominal 'levels' or should differences be used instead? I know with prices one needs to input the 'returns' (moves), but I wasn't sure about macro economic data levels versus moves.... Thanks for your thoughts!

2 responses

Hey Tanya,

It all depends on your specific hypothesis or what types of hypotheses you want the ML method to suggest. One of the main reasons you use returns over prices is that they're more likely to be stationary, but there are certainly times when having the prices as an input to the model with or without returns may make more sense. Likewise what preprocessing you do on the macro-economic data largely depends on what kinds of relationships you want to find with your ML model. Personally I like to use ML algorithms in research to find hypotheses and then test them out of sample. Then you can use the hypotheses as trading models once they're confirmed. You can also use them as online predictors but that tends to be a bit more finnicky. I would run the models in research with both the level data and differences as feature inputs and see which they like to use more.


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.

thanks! I think that's quite a neat explanation - like you said, sometimes we might want to feed in levels rather than differences... for example, I've read a few papers where they feed volatility levels rather than differences.