Quantopian's community platform is shutting down. Please read this post for more information and download your code.
Back to Community
Choosing the right ML Classifer
Zak Raicik
posted Oct 18, 2018
Machine Learning Statistics

Potentially a discussion too broad for the forums, so apologies if so.

I am working on using some type of ML to combine various alpha factors in one powerful alpha factor.

I am new to ML, so to make things simple- I chose to binarize returns. As a result, the goal of the ML becomes classification (as opposed to regression).

My question to the group is what different classification techniques should I be considering? From what I can find, the most common is the adaboost classifier but are there others I should be considering?

3 responses
Thomas Wiecki
Oct 18, 2018

If you want linear: ElasticNet
If you want non-linear: Random Forest and Gradient Boosting

Disclaimer

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.

Zak Raicik
Oct 18, 2018

Thanks. When evaluating what classifier works best for my purpose, what types of things should I consider(Size, accuracy, etc)? What classifiers are preferred for different situations?

I found this chart to be somewhat helpful...

http://scikit-learn.org/stable/tutorial/machine_learning_map/index.html

Thomas Wiecki
Oct 22, 2018

Zak: I would use the hold-out AUC as your criterion to select which classifier to use. If you also want to parameter tweak, I would run cross-validation on the training set. Lopez de Prado has a great new book on this you might want to check out.

About Quantopian
Careers
Community
Events
Documentation
Academia
Lectures
Status
Facebook
Twitter
LinkedIn
YouTube
Terms of Use
Privacy Policy

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.