Approximations And Abstractions
By now, most regular readers know how I feel about macro aggregates: They're largely meaningless.
Every day, policymakers of all kinds and capital allocators of all shapes and sizes consult statistical abstractions to make important (sometimes momentous) decisions with far-reaching consequences for other people's lives and money. In some cases, those abstractions are later revised so as to bear only a vague resemblance to the "data" as originally reported.
To be sure, approximations are the be
H
I really appreciate your continuing exposure of the fallacies that plague attempts at statistical forecasting and ridiculous attempts to create four and five digit statistical significance where it can’t exist. Volcker was right, of course. Having our lives ruled by people who are amateurs of questionable ability and understand little or nothing about the tools they are fooling (themselves) with is really annoying. It has always bothered me that we don’t do a better job at capturing the essence of individual behavior with something better than market metrics. As you have pointed out frequently, markets don’t buy and sell things, individual people (and gifted robots) do. Knowing what actually moves market participants remains largely unexplained. We have compensated by inventing techniques and devices we can bet on instead.