Quantopian's community platform is shutting down. Please read this post for more information and download your code.
Back to Community
Machine Learning: Turn $10K into $2.25M in two years (+22407% returns) by trading BRK_A (Berkshire Hathaway) with Random Forest
Upper GI
edited May 8, 2014

This is extending the framework proposed by JohnFawcett and Alex Izydorczyk about 2 years ago to use classification for predicting and acting upon trends based upon momentum.

Using historical returns as input into Random Forest training model. For each date, use the directional movements from each of the past 5 days as features to train the model. There will be 60 examples, which are simply the last 60 sets of trailing 5-day movements. Classification will predict as target the direction of movement of the stock on the present day.

For the current day, we use the trailing 5-day movements of the current date as features in the predictioon of the current day. Buy (long 10 shares) if the classifier predicts upward movement on the current date, sell (short 10 shares) if otherwise.

Results are shown for the algo run on BRK_A

2 responses
Dan Dunn
May 8, 2014

I added one line to your algo and re-ran the backtest. The line records your cash balance.

    record(cash=context.portfolio.cash)

Unfortunately, it looks like your algo borrows an outrageous amount of money. It would be cool to borrow $17,000,000 on $10,000 base. . . . but my broker won't let me!

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.

John Fawcett
May 9, 2014

Hi,

It is really fun to think about pairs of securities that are indicators or contra-indicators, so I started playing with this algo. I changed the algorithm to use target percentage orders instead of share count orders, and defined a step-size for going up and down in exposure to the stock. I also set a max long exposure and a min short exposure. The algo records the target percentage for the stock, so you can see the algo's sentiment over the course of the backtest.

The original shared algo was using BRK_A to indicate for BRK_A. To make that work with my percent orders, I had to start with at least $1M, since BRK_A has a share price in the six figures.

But, I found it more interesting to think about contra indicators. The attached backtest is using the SPDR Select Energy ETF as a contra indicator for the SPDR Select Materials ETF. My reasoning was that profitability in basic materials is inverse to that of energy because the manufacturing of materials is energy intensive.

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.

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.