Spurious correlations

16Sep06

A topic bubbling up in the blogosphere (and on this blog) is the importance of generating testable hypotheses about the capital markets. There is no shortage of opinions about how markets work, but only when they are translated into a form where they can be verified (or not) can they pass from the realm of conjecture to fact.

Those investors with a passing knowledge of the scientific method realize that science progresses in large part when existing theories are invalidated. Sharon Begley in the Wall Street Journal reported on the rise in scientific journals “…that publish only negative results…” This movement is a direct attempt to try and push back against the so-called “publication bias.”

“You hear stories about negative studies getting stuck in a file drawer, but rigorous analyses also support the suspicion that journals are biased in favor of positive studies,” says David Lehrer of the University of Helsinki, who is spearheading the new Journal of Spurious Correlations.

“Positive” means those showing that some intervention had an effect, that some gene is linked to a disease — or, more broadly, that one thing is connected to another in a way that can’t be explained by random chance. A 1999 analysis found that the percentage of positive studies in some fields routinely tops 90%. That is statistically implausible, suggesting that negative results are being deep-sixed. As a result, “what we read in the journals may bear only the slightest resemblance” to reality, concluded Lee Sigelman of George Washington University.[...]

Keeping a lid on negative results wastes time and money…

While the focus of these new publications is in science and medicine one could easily make the case that an analogous publication in the investment world would be of great service to investors of all stripes. Brett Steenbarger at TraderFeed has a piece up on how market blogs can help serve the process of theory testing.

Theories are our explanations of the world we see. When we observe and describe our sample of the world, we build a model based on our perceptions and say to ourselves, “This is what I think the world is like.” That model–our theory–provides us with the hypotheses that we test. When we check out and revise our hypotheses based on those tests, we’re really refining our internal models of the world: our theories.

In short, for Dr. Brett the best market blogs do more than talk about the markets, “..they model how to think about the markets.” Jeff Miller at A Dash of Insight also weighs in this question of hypothesis testing. For Miller listening to how investors make decisions is far more important than focusing on the inevitable noise that comes with short-term results.

In a previous post we used a discussion of string theory to discuss the importance of having testable hypotheses. As a follow-up we will point you to a piece by Gregg Easterbrook at Slate.com on the growing string theory debate. To the uninitiated string theory has been the dominant movement in theoretical physics for some time now. However to-date string theory is bereft of testable hypotheses, so argues a new book.

But string theory works only if you assume the existence of other dimensions—nine, 11, or 25 of them, depending on your flavor of string thinking—and there’s not one shred of evidence other dimensions exist. This may render string theory highfalutin nonsense that has hijacked academic physics.

It is important to keep an open mind on potentially world-changing theories. However we must also be on guard against the tyranny of “spurious correlations” that inhabit the investment world with regularity. This is of course, more difficult than simply affirming a positive belief in one’s world view, but nobody ever said this investing thing was going to be easy.

About these ads


Follow

Get every new post delivered to your Inbox.

%d bloggers like this: