School children with larger feet score higher on standard tests.
This is an old example in science and statistics, but if you’re not familiar with it, know this: It’s true. If you were to make a scatterplot of shoe size vs. test scores, there would be a very clear positive correlation. So does that mean that big feet cause big brains? Or vice-versa?
Well, no. The actual reason that these two variables are correlated with each other is that they’re both related to a third variable — age. Oh! Well…duh! Right, older students do better on a standard test because they know more, and their feet are generally bigger too.
So it’s important not to confuse correlation with causation. Here at The Capitalist Trader, my policy is to always look for some kind of causal mechanism to explain the relationships found in my statistical research. That way, we’re not just data-mining and coming up with statistical artifacts with no basis in the actual mechanics of the market.
The Relative Buying Power tool is a good example of this. As I explained in the last post, this tool is based on my hypothesis that the FOREX market tends to revert to the mean more often than it trends. The causal mechanism for this acts through the economic and political interests of the players, as I illustrated with a story.
The Weekly Price Projection tool is a case where the causal mechanism is not very well understood, but is still hypothesized to exist. The idea here is that significant price action, such as a breakthrough of the previous week’s high, will have an emotive effect on market participants who will then respond in some typical way due to trading psychology. Yep, that’s a bit hazy, but it’s a causal mechanism nevertheless.
So just remember, correlation is not causality and if your kids are getting really good grades then buy them some new shoes! Their feet probably hurt! Other than that, just…keep pipping up!!