As all students of economics know, time is limited. Given that, there has to be better ways to optimize its utilization than spending hours and hours working through or constructing irrelevant “New Keynesian” RBC and DSGE macroeconomic models. I would rather recommend my students allocating their time into constructing better, real and relevant macroeconomic models – models that really help us to explain and understand reality.Read it at Lars P. Syll's Blog
Wren-Lewis drivelling on macroeconomics
Lars P. Syll | Professor, Malmo University
Smack down.
1 comment:
Tom -- When I was at Wharton in the early 70's, econometrics was getting to be THE hot topic. With the new and rapidly expanding capacity of computers, this was seen as the answer to all the objections around measurement. With enough computing power, econometrics would finally be able to handle all the necessary variables to predict the movement of the economy or its major components.
Ad hoc conditions or limitations? No problem, we could handle hundreds, even thousands of variables. Add 'em in as we go. We could even build in the expectations effect and give separate predictions to the public and to the policy makers.
The golden age was upon us!
Except...it didn't happen. Econometrics will not be a very good tool for policy by itself. Even with tens of thousands of variables or millions. Rodger Mitchell has called for a computer center for economic research like the one being built for NSA to process all of our communications, but if we don't use the correct criteria in deciding which data is important, and how it may be interrelated, we won't get any useful information from such an effort.
People ARE irrational and driven by emotions. The markets have never been, and will never be truly "free". We need to get past that argument and deal with reality.
Post a Comment