Trading Twitter: Where Noise Becomes Signal

Over four years ago when we discussed the high frequency predator traders feasting on the signals of others, few believed it possible (and fewer still comprehended it). Today there is another potential disruptor in US equity market microstructure, the transformation of noise to signal from the overwhelming drivel of a Twitter stream. Macro signals (the hashcrash in April when AP’s account was hacked and this week’s Israeli military tweet misunderstanding) have had dramatic effects on the market but individual stock trading success remains elusive.

For social media stocks trading, “sentiment” implied from the Twitter stream leads prices by hours according to a recent study, but while tweets can contain useful information – it is far from guaranteed. “You have to be happy with a lot of noise in your data,” one advocate notes, but, as the FT notes, a recent PhD study of the translation of twitter noise to actionable signal perhaps sums it up best, “The proponents of this idea really do exaggerate it… I’m not saying there’s nothing here, but I’m not saying you can print money either.”

Of course, the biggest drawback (for now), not even the most advanced language processing algo can comprehend sarcasm.

Via FT,

traders are starting to look at Twitter not as a buy or sell in itself but as a way to generate hot tips on other stocks.

 

The idea is a simple one: the 500m daily messages on Twitter put online the sort of gossip usually only available in snippets by eavesdropping in bars or the office lift. Apply some moderately sophisticated computer filtering, and out should pop market-moving news and views investors can use.

 

Twitter has already demonstrated its potential to move markets, with the “hash crash” – named for the “#” symbols used on the microblogging service – knocking 145 points off the Dow Jones Industrial Average in April on the back of a false tweet from a hacked wire service account.

 

 

This week brought another example, with some traders blaming a misunderstood Israeli military tweet referring to bombing Syria in the 1973 war for a $1 per barrel drop in the oil price on Thursday.

 

 

“You have to be happy with a lot of noise in your data,”

 

 

The latest attempt to prove that Twitter contains useful information comes from Ilya Zheludev, a PhD student at University College London… “The proponents of this idea really do exaggerate it,” Mr Zheludev says. “If it really was as possible as they think it is, first of all they wouldn’t publicise it and second they would be rolling in cash.”

 

“I’m not saying there’s nothing here, but I’m not saying you can print money either.”

 

 

For most stocks there were not enough tweets to generate statistically significant volumes, while for the FTSE and many large companies – including Google, Intel and Bank of America – the sentiment had a stronger link to past price moves than future ones.

 

But 11 of the 50 financial instruments tested showed a statistically significant link,

 

 

 

Joe Gits, founder of Chicago-based Social Media Analytics, says that he expects a turnover of $60m in two years from his system.

 

 

He has big expectations. “I think it’s going to become every bit as big as earnings estimates in the not-too-distant future,” he says.

 

 

Previous attempts to trade purely on the back of social media data have come unstuck. Mr Peterson ran a hedge fund based on signals from social media, but gave up after finding it did not work for the biotech sector. In London Derwent Capital set up the first Twitter hedge fund in 2011, but quietly closed down again just a month later.

 

 

Mr Zheludev’s approach is rather more complex than the simple algorithms used by some early traders to extract signals from the online noise. But even the most advanced language processing has trouble with much of Twitter: they cannot process sarcasm.

 

Here is Ilya Zheludev presenting his thoughts last year at a TED Talk:

Social media is more than just Facebook updates and Twitter posts. The term also refers to any user-generated online content, from readers’ comments on articles in the Financial Times to blog posts. Social media is traditionally seen as just benefiting the end-user: the friends exchanging photos on Facebook. But its value is far greater than that — the amount of publically accessible data that social media generates is phenomenally large. Just by way of example, there are 6 million Facebook views every minute. Harnessing that largely untouched data correctly has unbelievable implications: we’re able to see into the future. We can predict so many facets of life, from the spread of disease to the time and location of the next military uprising. Social media lets us see into the minds of millions, in real time — something that has never been possible before. The future isn’t a great unknown any longer.

 

Being specifically interested in Finance, Ilya’s research is driven by the relationship between the internet’s ‘mood’ and financial markets.

 

 

The bottom line is that in a world in which signal is gone courtesy of central planning and where only the Fed’s H.4.1 statement (the weekly balance sheet update) matters (as we said back in January 2010); market participants are desperate for any “edge”, even if it means creating signal out of pure reactionary noise.

    



Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.