Quantopian's community platform is shutting down. Please read this post for more information and download your code.
Back to Community
Access Twitter and StockTwits Trader Mood data
Seong Lee
edited
Using Data Sentiment

How to access Twitter and StockTwits Trader Mood data

PsychSignal provides Social Media Trader Mood sourced from both StockTwits and Twitter and you'll be able to use the data in your algos.

For a full summary here are the datasets that you can use:

Examples

Here's some basic examples on PsychSignal for you to use as a starting point or clone the sample algorithm below:

Our allocation process attaches high value to algorithms that use a wide variety of datasets. We evaluate all algorithms that use alternative data, including strategies that use either free datasets or premium datasets. And, this dataset in particular does not require a subscription, which means you can research and backtest using the full dataset.

Happy Coding!

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.

18 responses
Seong Lee

The research for the sample algorithm above is found here: https://www.quantopian.com/posts/work-in-progress-social-media-message-volume-as-a-proxy-for-stock-volatility

Martin Clasen

Great Deal!
Your algo looks awesome, also it's a good example of pipeline usage and good practices :)

Sometimes, it feels like we have to learn something new every day to make a well-made strategy, but we know that it is the result of many iterations finding the evolution of the platform: easy, reliable and powerful.

Q2/Pipeline is amazing!

Thanks,

Deleted User

@ Seong - Is there any way to get a better understanding of how StockTwits generates their bearish signals? Would love to better understand the science behind the data.

Also, Awesome idea! I am going to start playing around with the data to test the notion that bearish sentiment is more powerful than bullish sentiment. The idea comes from the monkey study found in the new Netflix documentary "Boom Bust Boom". Maybe StockTwits aggregate bearish data can provide some insight into a possible stampeding of the heard.

For now, just wondering if anyone thinks it is odd that bearish sentiment for Apple is relatively low this year? I would have thought that the first ever quarter of revenue decline in the Iphone era would have boosted bearish sentiment. I bet the data needs some filtering for macro factors and such.

Peter Bakker

awesome, really great to be able to play with it live and direct

Josh Payne

@Frank, you might want to listen to this webinar we did a few months back in collaboration with PsychSignal: https://www.youtube.com/watch?v=Ul5LPjkuUEA

If memory serves, James (CEO of PsychSignal) gave some background on their methodology.

Thanks
Josh

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.

Deleted User

Thanks as usual Josh! Will give it a watch tonight.

Deleted User

Josh or Seong,

I have begin looking at the PyschSignal data in both Research and the IDE. One quick question...Is there a way to limit the data to specific stocks in the backtester without the use of pipeline? It appears easy enough to use a PyschSignal factor with my pipeline output, but what if I just want to restrict the data to a specific security (say APPL) without having to use a pipeline? A simple example would be to just have my algo print the bullish sentiment score every day for Apple.

Thanks

Seong Lee

Frank,

That's a great question.

I'm thinking you could do something like:

pipe.add(psychsignal.bullish_intensity.latest, 'bullish_intensity')

and later get the values for that through:

output = pipeline_output('bullish_intensity'); output.ix[sid(24)]['bullish_intensity']

Deleted User

Looks good Seong. I will give it a try and let you know if I have an issue.

Thanks!

James Hutchison

I find the StockTwits name very appropriate, since a lot of the people on it are indeed twits.

Be sure to filter out ETFs. A lot of people will tweet a series of related ETFs in their tweet and stocktwits will get confused and stick "bullish" on one of them.

Does this go by unique messages from unique people, or just messages? Seems like you'd want unique individuals

Josh Payne

Hi James,

It's important to realize that, for this particular data set, the analysis of bullishness and bearishness is not using StockTwits raw data -- PsychSignal is analyzing the text of the message from StockTwits and assigning their own assessment.

Check out some of the field descriptions: https://www.quantopian.com/data/psychsignal/stocktwits

Thanks
Josh

Seong Lee

Here's an algorithm shared by Paul Cao: https://www.quantopian.com/posts/xiv-slash-vxx-pair-trade-1

rakhi batra

Can I get textual data(Tweets) of StockTwits by specifying symbol and time period?

Honver Lam

The 1 year limit is coming up soon. What will be the monthly cost for this dataset after that?

Seong Lee

Thanks for the question Honver,

Barring any other changes -- the PsychSignal dataset will be available for use in your algorithms.

As a note, our allocation process attaches high value to algorithms that use a wide variety of datasets. We evaluate all algorithms that use alternative data, including strategies that use either free datasets or premium datasets. And, this dataset in particular does not require a subscription, which means you can research and backtest using the full dataset, as well as paper and live trade algorithms that use it.

Here's a good example of another algo using StockTwits & PsychSignal

Robbert

Hi im new here,

Is there a possibility to download the raw data (provided by PsychSignal) to my computer?
I want to do some regression analyses on the stock data, twitter data signals and some other factors.

Kind regards/

Ali Bilgic

Good evening. I have a Problem with backtesting the sample algorithim ofTwitter and StockTwits Trader Mood (All fields, with Retweets).
The backtest shows no returns, only the Benchmark line? Can please somebody help me out of it, it is necessary.
Thank you

Miltiades Chryssavgis

Thanks for sharing this, interesting data to analyse and potentially incorporate into strategies.

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.