- Title Pages
- Preface to the New Edition
- 1 Econometrics – Alchemy or Science?
- 2 Stochastic Specification in an Aggregate Demand Model of the United Kingdom
- 3 Testing Dynamic Specification in Small Simultaneous Systems: An Application to a Model of Building Society Behaviour in the United Kingdom
- 4 Dynamic Specification
- Part II The Development of Empirical Modelling Strategies
- 5 On the Time‐Series Approach to Econometric Model Building
- 6 Serial Correlation as a Convenient Simplification, not a Nuisance: A Comment on a Study of the Demand for Money by the Bank of England
- 7 An Empirical Application and Monte Carlo Analysis of Tests of Dynamic Specification
- 8 Econometric Modelling of the Aggregate Time‐Series Relationship Between Consumers' Expenditure and Income in the United Kingdom
- 9 Liquidity and Inflation Effects on Consumers' Expenditure
- 10 Interpreting Econometric Evidence: The Behaviour of Consumers' Expenditure in the United Kingdom
- 11 Predictive Failure and Econometric Modelling in Macroeconomics: The Transactions Demand for Money
- 12 Monetary Economic Myth and Econometric Reality
- Part III Formalization
- 13 The Structure of Simultaneous Equations Estimators
- 14 AUTOREG: A Computer Program Library for Dynamic Econometric Models With Autoregressive Errors
- 15 Exogeneity
- 16 On the Formulation of Empirical Models in Dynamic Econometrics
- 17 The Econometric Analysis of Economic Time Series
- Part IV Retrospect and Prospect
- 18 Econometric Modelling: The ‘Consumption Function’ In Retrospect
- 19 Postscript: The Econometrics of PC‐GIVE
- 20 Epilogue: The Success of General‐To‐Specific Model Selection
- (p.330) 15 Exogeneity
- Econometrics: Alchemy or Science?
David F. Hendry (Contributor Webpage)
Robert F. Engle
- Oxford University Press
Exogenous variables play a crucial role in econometrics, yet ‘exogeneity’ is often imprecise. Exogenous connotes ‘being determined outside of (the model under analysis)’, so it cannot be a property of variables per se. Rather, exogeneity is a step in model reduction, concerning when inferences about parameters of interest based on a complete analysis of the joint density function of all the observable variables coincide with inferences based on only the conditional density of one sub‐set of variables given another sub‐set. If there is no loss of information from only analysing the conditional sub‐model, then the conditioning variables are weakly exogenous for its parameters of interest, and the marginal process is irrelevant. To forecast conditionally more than one period ahead also requires Granger non‐causality, leading to the concept of strong exogeneity. To justify conditional policy analyses, which change the marginal model, parameter invariance is required in the conditional model, leading to super exogeneity.
Oxford Scholarship Online requires a subscription or purchase to access the full text of books within the service. Public users can however freely search the site and view the abstracts and keywords for each book and chapter.
If you think you should have access to this title, please contact your librarian.