Wednesday, September 17, 2008

Gut instinct and quant prowess may not be opposites

The general view is that those who have good quantitative skills are not the same people as those who use gut instinct or have better ability at gaining sense of relative magnitude or the ability to size things up. Intuitive counting is not the same as computational logic.

Recent research on learning behavior shows that there may actually be a connection with the primitive cognitive skill of making quick number judgments and the refined skills of computational math. These two maths systems, bestial or primitive versus celestestial or formal may be closely correlated. Those who have good primitive skills have shown to have good computational skills. This dichotomy is nicely described in a recent NYT science article by writer Natalie Angier,http://www.nytimes.com/2008/09/16/science/16angi.html?partner=rssnyt&emc=rss.

In some sense, better mathematician should be better at the simple non-symbolic skills of approximating. You can count trees but you also have to have some skill at approximating the size of the forest. Scientist do not know how our crude approximate skills are tied with the skills to do higher math or whether you can be trained in one area and improve in the other. Nevertheless there may be innate math or number skills which transcend both the high and low level versions of counting. Finding a ballpark figure, a Fermi problem solution, is a good skill to have and helps with other math.

I like to believe that intuition in thinking about approximate magnitude is an important skill for any investor which helps with computation skills. You need to have a good feel for numbers to be effective. For example, in forecasting the precision of the forecast i not as important as getting the direction right. Working on these directional skills is important as a first start before plowing into the specifics of econometric modeling. Knowing the trend is as important the precise number.

So thinking about how many Starbucks are there per person in the US is as important as estimating the expected earnings of the company. Thinking about the number of cars is China is relevant for forecasting the price of oil.

No comments: