Date of Source: 12 Apr 2016
As quants, we’re all aware that every model has a shelf-life… a similar pattern applies to the world of data. Rare, unique and proprietary data eventually diffuses and becomes commonplace, easily available, edgeless data. The best analysts constantly reinvent their models, to avoid their inevitable obsolescence. Today, they’re venturing into the world of alternative data as a new source of alpha. […]
Quandl founder Tammer Kamel concisely describes the process which underlies “alpha decay”, i.e., how historically predictive trends lose their predictive power, even when a causal relationship can be ascribed. The phenomenon of alpha decay is probably the foremost factor which causes quant traders and funds to fail. In an efficient marketplace, the assumption that the future looks like the past can only be true when the market participants have no knowledge of the patterns which describe (ascribe?) their behaviors. This is like Heisenberg’s Uncertainty Principle. It is also an underlying precept of psycho-history; a concept pioneered in Isaac Asimov’s Foundation series.
Tammer prescribes the adoption of alternative data sets as a means to combat unavoidable alpha-decay. I get it — unique information and such. However, he does not address the persistence of specific market-based phenomena that, despite the market’s long standing knowledge thereof, continue to persist in statistically significant and measurable ways. Two prominent examples are value and momentum.
The continued preeminence of these phenomena may, in fact, be rooted in a single cause: they are both premised in taking positions contrary to the market consensus. I contend that being a contrarian works because it exploits a behavioral complex which is deeply embedded in human biology: humans persistently under-react to good news and over-react to bad news. Moreover, being a contrarian tends to have a positively skewed reward-to-risk profile: the pay-offs for taking positions which contrast with the consensus tend to be asymmetric, even if the market is usually right.
So, alternative data… I see the value, but only if one has already exhausted all the low-hanging fruits.
Blog Post Source: Tammer Kamel. Alternative Data – The Newest Trend in Financial Data. Quandl. 12 Apr 2016. Original content accessible: https://blog.quandl.com/alternative-data
Read also on the Quandl Blog: The Unbearable Transience of Alpha
Updates to the Read List are meant to keep readers informed of significant news stories and analytical highlights from third party sources. Content that was recommended prior to 2017 is accessible through the Read List Archives.