Econometric Modelling: The ‘Consumption Function’ In Retrospect

David F. Hendry

in Econometrics: Alchemy or Science?

Published in print October 2000 | ISBN: 9780198293545
Published online November 2003 | e-ISBN: 9780191596391 | DOI:
 Econometric Modelling: The ‘Consumption Function’ In Retrospect

Show Summary Details


Exposits the main ideas and concepts for consumers’ expenditure behaviour in the UK to clarify their empirical content and implications. Each theoretical concept in the information taxonomy is related to the relevant empirical counterpart, reinterpreting some of the intuitive notions in DHSY, to show how the formalization clarifies problems with a ‘conventional’ econometric approach. The discovery/evaluation dichotomy, and the necessity but insufficiency of evaluation criteria to determine validity, are stressed. Modelling is an exercise in design to achieve predetermined selection criteria, but rigorous evaluation can eliminate poor designs. The theory of reduction explains the derivation of empirical models and their susceptibility to design, leading to an analysis of design and conventional criteria. The six information sets of the taxonomy and their associated criteria are related to well‐known test statistics (where each criterion is the null). The resulting tests are applied to the consumers’ expenditure model in DHSY on quarterly and annual data frequencies, and the inter‐war and post‐war UK data are compared, illustrating the progressive research strategy that follows from an encompassing approach to econometrics.

Keywords: consumer expenditure; DHSY; encompassing; evaluation criteria; general‐to‐specific modelling; information taxonomy; model design; progressive research strategy; theory of reduction; UK

Chapter.  12587 words.  Illustrated.

Subjects: Econometrics and Mathematical Economics

Full text: subscription required

How to subscribe Recommend to my Librarian

Buy this work at Oxford University Press »

Users without a subscription are not able to see the full content. Please, subscribe or login to access all content.