NobleBlocks

Research indexBook chapter

Comment

Quah

15 Dec 1994

NobleIDark:/48914/w1/20260522/04CF9431DOI10.1093/oso/9780198288596.003.0017

Paywalled

The full text is at the publisher

You can read the abstract here, or open it at the publisher.

Read at the publisher

Abstract

Abstract In the minds of many economic theorists and many traditional econometricians, vector autoregression (VAR) analysis is one of the worst things that could have happened to the scientific research programme that is business-cycle study. Users of VARs commit every crime imaginable: they estimate reduced forms—and hardly ever calculate formal specification tests—instead, preferring to work with a fiction known as a just-identified model. These practitioners refer to things as structural that no econometrician—working in either standard simultaneous equations, standard time series, or standard rational expectations paradigms—would recognize to be structural. (Within those three groups, of course, none acknowledges any of the others as structural either.) No undergraduate econometrics course teaches the tokens of VAR research dialogue: impulse response functions and forecast error variance decompositions.

Discussion

Topics

More

Suggest a correction or report