Can Big Data predict stock price movements?
A few years ago, following the apparent discovery by the OPERA research group of neutrinos traveling faster than light from a source in Switzerland to a detector in Italy, lots of physicists set out to explain how this might occur. One paper by physicist Michael Berry and colleagues was entitled "Can apparent superluminal neutrino speeds be explained as a quantum weak measurement?" Their abstract was a model of concise efficiency, and I quote it below in full:
That's essentially also the conclusion of this new and completely unrelated paper by physicists Damien Challet and Ahmed Bel Hadj Ayed looking at whether Big Data in the form of information from Google Trends can be used to predict movements of asset prices. There have been a number of studies in recent years suggesting that this might be possible; that Big Data might be used for "nowcasting" of asset prices and go beyond what is possible with historical price data alone. Is it possible? These two authors conclude "probably not," though not in quite those words. Their findings, they write,
... can be summarized as follows: using this kind of data to predict volume or volatility is relatively easy, but the correlation with future price returns is much weaker. Incidentally, this matches the daily experience of practitioners in finance who use price returns instead of fancy big data.
Here we discuss what can go wrong in every step required to backtest a trading strategy based on GT data. We then use an industry-grade backtest system based on non-linear machine learning methods to show the near-equivalence of the exploitable information content between SVI and historical price returns. We therefore conclude that price returns and GT contain about the same amount of predictive information, at least with the methods we have used and challenge to community to do any better.
I can't say I'm surprised.
(Thanks to my colleague Justin Mullins for pointing out the "probably not" abstract to me).
Follow The Physics of Finance on Twitter: @Mark_Buchanan