I am utilizing the DecistionTreeClassifier predict_proba() function to generate an Alpha Signal. Since the classifier is one of two values (0, 1), I hit the max_loss threshold immediately when generating the get_clean_factor_and_forward_returns() and later factor_returns() functions. Any ideas on how to avoid this and properly test my classifier prediction? The end goal is to calculate the Sharpe Ratio of the Alpha Signal factor returns.
Thanks.
Hi Chris,
Predict should give you only 1 answer - in your case a 0 or 1... predict_proba will give you two probabilities (as you appear to be doing a binary classification from your description). The two values from predict_proba are the probability of a zero and the probability of a 1.
If you are using predict_proba and you only get back 0's and 1's then your data is perfectly seperable, which means the classifier is able to achieve 100% accuracy. As this probably isn't expected from your post description, I would look carefully at the data you are sending into the tree...
cheers,
Bruce
Bruce,
After taking some time to re-work several things, I am rounding back on this question but re-phrased instead. The following is hitting a 100% drop...get_clean_factor_and_forward_returns(factor=factor_data, prices=pricing_data, periods=[1]).
My factor_data is...0 or 1 for multiple years in a large stock universe. My pricing_data is simply pricing for the same multiple year timeframe plus 30 days.
Is the 100% drop due to the lack of different factor data values...every value is either 0 or 1?
Thank you,
Chris
Bruce,
I found a great post that gets to the bottom of this. Thanks for your help along the way.
https://www.quantopian.com/posts/how-can-i-use-alphalens-with-boolean-factor-true-and-false
Chris
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.
