Quantopian's community platform is shutting down. Please read this post for more information and download your code.
Back to Community
Discretizing a continuous variable to build a dataset for classification
Andrew Czeizler
posted Apr 27, 2015

From the papers I have read, it seems that it is the direction and strength of movement that is the most importance not necessarily the price itself.

My goal is to create a dataset that has features and has the predictive variable Bucketed, such as histogram. I would like to know once the variable has been bucketed using a method, how do I then get the buckets back and attach it to a dataset so that I can proceed with a classification method.

I could discretize the predictive variable using the Bayesian Blocks algorithm as shown below.

http://www.astroml.org/user_guide/density_estimation.html
or
https://github.com/pkgw/pwkit/blob/master/pwkit/bblocks.py

Many thanks in advance for the assistance.

2 responses
Casey Detrio
Apr 28, 2015

Bayesian Blocks! Hadn't heard of that, cool.

I don't understand your question.. how do you "get the buckets back"?

Are you talking about the programming environment? for importing data, running scripts, and so on? My preference is to center things around an ipython notebook (for example). Quantopian's website is centered around their web code editor. But their tools can be used with ipython notebooks, and the beta "research" platform goes in that direction.

Andrew Czeizler
Apr 29, 2015

Cheers Casey!
That is a very informative notebook.
By question just to clarify was, that as I understand it the Bayesian blocks algorithm automatically buckets a continuous variable by using
a change point. I assume this means that this buckets once it has gone through the entire continuous variable are stored some where as data.
I was wondering how to query theses buckets and then attach them as a column to a data set.
In the above notebook it states that the bins should be plotted but how do a print out the actual bin categories that create the histogram.

Many thanks
Regards,
Andrew.

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.