Further chatter on using social media ‘tweeting’ as predictive indicators in trading


Social media as predictive indicators for trading – coming to a screen near you soon.

Following on from Mike Hill’s post last month a number of banks we are speaking with are talking (or in social media speak, “tweeting”) about incorporating some form of social media sentiment and correlation analysis as a predictive indicator within their trading platforms – to uncover hidden trading opportunities.

Based on the research paper below, these positive comments and increased frequency of such ‘chatter’ are in themselves a good predictive indicator that banks will do just that (sort of obvious really).

Here is why it’s interesting. Below is an extract from a research paper which analysed the pre-release twitter chatter (positive and negative) for upcoming but yet to be released movies, and then showed that based on an analysis of positive/negative chatter, they were able to predict the winners/losers, and actually show a very high correlation against the Hollywood Stock Exchange index (HSX)


Using the rate of chatter from almost 3 million tweets from the popular site Twitter, we constructed a linear regression model for predicting box-office revenues of movies in advance of their release. We then showed that the results outperformed in accuracy those of the Hollywood Stock Exchange and that there is a strong correlation between the amount of attention a given topic has (in this case a forthcoming movie) and its ranking in the future.

We also analyzed the sentiments present in tweets and demonstrated their efficacy at improving predictions after a movie has released. While in this study we focused on the problem of predicting box office revenues of movies for the sake of having a clear metric of comparison with other methods, this method can be extended to a large panoply of topics, ranging from the future rating of products to agenda setting and election.

Now turn that same predictive analysis towards stocks, commodities, FX rates, debt markets. Imagine some form of twitter sentiment analysis, presented as dynamic ‘heatmaps’, where markets/sectors/instruments or currencies are shown grouped by twitter chatter coloured to show high frequency of positive/negative chatter sentiment – that’s of value to clients.

Full research paper here

4 Responses

  1. interesting stuff – check out this also https://www.recordedfuture.com/ – shame there’s no free trial, curious to see what it could throw up..

  2. Thank you for that, yes this is the sort of thing,
    Paul

  3. Mike,
    I just happened on this and your previous post – good stuff. I work in business development at Gnip; we aggregate and deliver social media data from around 40 sources (including one of the exclusive agreements to resell Twitter data) to clients across a range of industries.

    The last six months have been really interesting – we’ve seen a lot of outreach from the finance/trading world – mainly funds who are looking to make this a factor in their trading algorithm. If you decide to post on this again, feel free to hit me up (link to Twitter included) and I’m happy to share what I can (obviously confidentiality is pretty important to our clients!).
    Seth

  4. […] We have recently covered leveraging social media trends in trading. […]

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: