Forum Archive Index - November 2002

["Capitalist" <capitalist@paradise.net.nz>]

quote:It implies that his computer programs can predict the future, or human
actions, which is also a load of bollocks.

 

And this is the crux of the matter in statistical modelling. You cannot predict human action by mathematical means, though modelling can be useful in equilibrium like phases of markets. By its very nature statistical modelling abstracts out the most crucial part of economics - human action.

 

A quote from Mises in Human Action V.1

 

"Logic and mathematics deal with an ideal system of thought. The relations and implications of their system are coexistent and interdependent...Within such a system the notions of anteriority and consequence are metaphorical only..The system itself implies neither the category of time nor that of causality. There is functional correspondence between elements, but there is neither cause nor effect. What distinguished the praxeological system from the logical system is precisely that it implies the categories of both time and of causality."

 

As I frequently say - statistical predictions are futile without knowing their causal identity.