Can Twitter Really Drive Investment Decisions?

Image representing Twitter as depicted in Crun...

Image via CrunchBase

A group of hedge-fund managers are launching a multi-million dollar hedge fund next month, using Twitter as its market indicator to determine sentiment and to thereby make investment decisions.

This information came from a recent article on CNBC / Yahoo Finance which quoted Derwent Capital Markets – a London-based hedge fund – as saying it had successfully marketed the new venture, officially called the Derwent Absolute Return Fund, to high net-worth clients and had attracted over £25 million in investments.

The company is confident it can achieve returns of at least 15-20% per annum by analyzing information gathered from over 100 million tweets each day, which the firm brands as “The 4th Dimension.”

On the face of it, this may sound like a risky, or even crazy, venture – but is it?

Let’s face it, the concept of rational markets has been comprehensively debunked during the last few years of economic crisis, and the global growth in wealth came to a dramatic end largely through a change in general sentiment. We’ve also seen plenty of allegations – many apparently backed by evidence – of collusion between those in research and those in investment banking to pump stock prices of certain companies at various times. In fact, based on this and my own experience, it seems that relying on the “experts” to manage your investments is no greater guarantee of success than simply using a general market-tracking fund – and often provides worse returns.

Furthermore, most people agree that we won’t see real growth return this cycle until consumer confidence picks up. Isn’t that really just about general market sentiment?

So contrary to some of the views on this fund, I would argue that this is a smart bunch of people – what they’re doing is using current technology to gauge market sentiment and make investment decisions from there.  Instead of listening to a small group of people to try to understand what “the man in the street” is saying, they’re tapping into the collective feelings of millions.

I see this as the start of a whole new way of tapping into societal collective wisdom and sentiment. What do you think?


8 responses to “Can Twitter Really Drive Investment Decisions?

  1. Im surprised that this idea hasnt been attacked by more startups; especially in London.

  2. I’d like to see a simulation of what it would have done over the last few years, but it looks interesting.

    They seem to drink the whole Twitter firehose, analyse it for mood type things, namely:
    which gives them a 6 dimensional “mood spectrum”. Presumably they map this against particular stocks, and the mindset of those driving trades generally.

    Given that since at least the time that Soros got out of the market, “market sentiment” has been a major force, this makes some sense. Some would say that such sentiment is a fine example of the “idiocy of crowds” but they are not entirely correct even if that well describes crowds.

    Algorithmic trades and trades that get in on privileged market information (before the wires even carry it to the rest of humanity) skew the picture. These guys (the quants behind such trades) might be smart but they do some ultra dumb things too!

    Given that proviso, I think the approach might have real legs. It looks like a way of measuring the “ambient decision atmosphere” that I’ve noticed seems a part of how many humans think.

    There’s also the fact that much of your best input is from people who are often really busy. Why tweet before you make that killer trade!! maybe after! maybe you’ll be too busy for three days! So some of the people who are NOT MAKING TWEETS are being left out of the data processing. Maybe with the periodic catastrophic impact.

    Then there’s the Munich Traffic Forecast Phenomenum. When the traffic predictors for the autobahnen around Munich started they were super useful. Then everybody started using them, they changed their behaviour as a result, the data became much worse. Predicting the market, if it works will have the same impact. Even if these guys play it close, if it works, somebody will come up with a publicly available (open source / smart phone app / web site / “we tweet it”) version which will skew it all with feedback.

    • Some great input, thanks Mike. My much more simplistic view is simply that markets are driven largely by public sentiment anyway, and that this just provides a potentialy more immediate way of gauging this. Of course, you can – as you say – get into the self-fulfilling prophecy or self-destructive loop thing, too. We’ll have to see how this plays out but I expect social media to play an increasing role in this sort of “collective wisdom” data gathering/analysis.

      • Reading my post again I see that it says that the quants do dumb things. That’s not what I meant. I think the quants understand what they’re doing but it’s their management that is guilty of simplified thinking.

        I see a parallel with the colleralised debt / mortgage scam that gave us a GFC. If the people who manage this system don’t understand it’s limitations and it gets popular we could see another “falling off the cliff”.

        I just found a nice article about the mathematical underpinings of the GFC at the Economist. (Li is the guy who published details of using a mathematical technique called a Gaussian Copula to estimate hitherto unknown things in some very powerful financial equations.)

        On August 10 2004, however, the rating agency Moody’s incorporated Li’s Gaussian copula default function formula into its rating methodology for collateralised debt obligations, the structured finance instruments that subsequently proved the nemesis of so many banks. … A week after Moody’s, the world’s other large rating agency, Standard & Poor’s, changed its methodology, too.

        The big and fatal change that occured after 2001 … If they did this as a result of the Gaussian copula, it screams incompetence. The same can be said for rating agencies who should have known better.

        Like any elegant model, the Gaussian copula has some value. If you appreciate the local conditions, it provides a convenient way to describe a complex relationship. But like many useful innovations, when it is exploited by the wrong people it can cause great harm.

        One day maybe you can insert POMS-Twitter-Analysis for Gaussian copula in a similar article!

        I also suspect that if Twitter is all that they’re using, then they lack some ambition and breadth of vision. (Given what I’ve seen of their press release they may indeed have painted themselves into that corner!)

  3. Smart idea – time will tell if it works. Not sure – maybe, maybe not.

    • Thanks Catarina – as Mike says, there are potential pitfalls but given that markets move on public sentiment anyway, I am of the opinion that this is more likely to be successful than not.

  4. Pingback: Tweets that mention Can Twitter Really Drive Investment Decisions? | Whit's End --

Leave a Reply

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

You are commenting using your 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