Re: Utility maximizing

[Peter Dorman <dorman@xxxxxxxxxxxxx>]

I think the most important inference to be drawn from current debates
over choice theory is that there is no obviously "correct" level of
analysis.  If we had a coherent, empirically validated theory of
individual behavior at the level of individual choice, then we might say
that this is the preferred level of analysis, and we would only employ
macro analysis when practical problems of complexity or data
availability intruded.  But we do *not* have such a theory, and there is
no corresponding general preference for individual-level analysis.  For
some purposes (perhaps the study of addiction), the appropriate level of
analysis is below the level of the individual (multiple preference
orderings, etc.), for others it is at the individual level, and for
others it is above, i.e. macro.  This is fundamental, in my opinion.

Peter Dorman