So in theory I am working towards a larger exposition on this issue, but as I get the chance I will shoot out various notes. Barry Ritholtz gives me such a chance.

In other words, yes, we can figure out actual causation.

Where I part ways with Hume is in looking at causation analysis as merely competing stories tying two facts together. It is larger than that, there is Causation-in-Fact. At the very least, we can eliminate the narratives that are demonstrably false. But to do so, we need to avoid the over simplifications, the correlation errors, the misapplied data, and recognize the complexity of causation in the world of finance.

Suppose you had figured out actual causation. How would you know?

About these ads