 |
|
|
Links to Research Summary Pages
|
|
|
|
|
|
|
|
|
|
|
|
To open, downlaod or print a copy of the full document please select the following link:
Our approach to building investment portfolios is focused on scientific / academic research. We believe this is crucial as this research has stood up to the rigours of a peer review process whereby it is assessed for its quality before being published. This scrutiny works to ensure that the research can be relied upon when making important investment decisions.
Stupid Data Miner Tricks: Overfitting the S&P 500
The Journal of Investing
David J. Leinweber (Haas School of Business, Berkeley, University of California)
Spring 2007
What the paper says: It provides examples of blatant, totally bogus applications of data mining in finance to make the point of the need to be aware of the risks of data mining in quantitative investing. The examples showed that the production of butter in Bangladesh explained 75% of the variation in US stockmarket returns. The production of butter in Bangladesh and the US; the sheep population in Bangladesh and the US; and the production of cheese in the US explained 99% of the variation in US stockmarket returns (which is a an extraordinarily high amount of explanatory power)
In the author's own words: "The dark side of data mining is to pick and choose from a large set of data to try to explain a small one . When data mining techniques are used to scour a vast selection of data to explain a small piece of financial market history, the results are often ridiculous. These ridiculous results fall into two categories: those that are taken seriously, and those that are regarded as totally bogus. Human nature being what it is, people often differ on which category is which."
The following links highlight the key scientific / academic research (not data mining) we have used to build portfolios for our clients:
|
|
|
|
|
Adobe Reader® required
