Sunday, May 21, 2017

Phil Price — An obvious (?) fact about constrained systems.


In economics the constraints are restrictive assumptions introduced for tractability and simplification.

The post makes the point that when one constraint is relaxed in a precise way in that can be measured with respect to the system, the response of the system is knowable. This is how economic models are used.

However, if all the constraints imposed by the restrictive assumptions of economic models are removed, as they are in the real world, the behavior of the actual system becomes unknowable from the model, a point that Keynes noticed and pointed out.

This is a reason it is difficult to develop models that work as representational models where human motivation and behavior is involved. Such models are infected with uncertainty. As a financial speculator, Keynes was well aware of this as a person that had written a book on probability theory.

As Richard Feyman said, "Imagine how much harder physics would be if electrons had feelings!"

Statistical Modeling, Causal Inference, and Social Science
An obvious (?) fact about constrained systems.
Phil Price

1 comment:

Ryan Harris said...
This comment has been removed by the author.