Thursday, February 18, 2010

Macro economic modeling works badly unless

we are severely constrained in which case the macro modeler can reduce the number of factors. This is in response to a quote of Kling's, to wit:

An observational study can be of scientific use if the conditions are right. One condition is that there are many observations relative to the number of factors that must be controlled for. In statistical jargon, this is known as the degrees of freedom.
But Kling's Recalculation must involve the ability of agents to perform some macroeconomics. Agents can accomplish this without direction because agents are faced with few, or perhaps one, severe constraint (in a severe recession). Hence macroeconomics devolves to solving that particular constraint.

No comments: