From the Archives: Why Systematic Models Are Great

James Montier wrote this piece in 2006, but it is so great that I have to bring it up again!  This article is a gem, worth reading over and over again.

What could baseball, wine pricing, medical diagnosis, university admissions, criminal recidivism and I have in common? They are examples of simple quant models consistently outperforming so-called experts. Why should financial markets be any different? So why aren’t there more quant funds? Hubristic self belief, self-serving bias and inertia combine to maintain the status quo.

Montier gives numerous examples of situations in which the models outperform both experts and experts using the models as additional input.  Using your “expert knowledge” just makes it worse most of the time.  In fact, in a study of over 130 papers comparing systematic models with human decision-making, the models won out in 122 events.

So why don’t we see more quant funds in the market? The first reason is overconfidence. We all think we can add something to a quant model. However, the quant model has the advantage of a known error rate, whilst our own error rate remains unknown. Secondly, self-serving bias kicks in, after all what a mess our industry would look if 18 out of every 20 of us were replaced by computers. Thirdly, inertia plays a part. It is hard to imagine a large fund management firm turning around and scrapping most of the process they have used for the last 20 years. Finally, quant is often a much harder sell, terms like ‘black box’ get bandied around, and consultants may question why they are employing you at all, if ‘all’ you do is turn up and crank the handle of the model. It is for reasons like these that quant investing will remain a fringe activity, no matter how successful it may be.

Lack of competition may be the best reason of all to use a systematic approach.  How many investors are willing to go through a thorough and rigorous testing process to build a robust model—and are then willing to stick with the model through thick and thin?  As Montier points out, it may remain a “fringe activity” no matter how successful it is.

—-this article originally appeared 12/22/2009.  This is a powerful, powerful argument in favor of using a systematic model.  Montier’s discussion of why investors resist using models is still very true.

Comments are closed.