Econometrics (any form of data crunching) is open to all kinds of charlatans, but so is theory, history of thought, and economic history and philosophy. Many economic relationships may be inherently nonlinear. I'd certainly believe this to be the case. But, there are non-linear estimation techniques. As I've stated before, there never has been and there never will be a flawless econometric equation. But, for any group of analysts who are in reasonable theoretical agreement, it is not difficult to assess the relevance of a particular econometric studies or collection of studies. The problem occurs when there are theoretical differences. peace, patrick At 10:47 AM 11/21/01 -0800, you wrote: >With econometrics, do data constrain speculation or does speculation >constrain data? Here Patrick, Allin and everyone else on this list would >be able to speak to the problems infinitely better than I can. As Wolfgang >Stolper argues, non-linear equations of high degree have multiple >solutions, so isn't it the econometrician who is making decisions all the >time concerning the algorithm the computer is going to follow? For a >layperson like myself, this certainly goes against the image of the >econometrician who tries to determine the starting point as quickly as >possible while letting the computer find the answer(s) without trying to >influence further the results of the calculations which the computer >makes. And with new data becoming available all the time, the >econometrician has to make some estimate of some short term developments >in feeding data to the computer. So why doesn't this open up econometrics >to all kinds of charlatans, as WS fears? >Rakesh
This archive was generated by hypermail 2b30 : Sun Dec 02 2001 - 00:00:06 EST