Econometrics Alchemy or Science? Essays in Econometric Methodology

by Hendry, David F.

• ISBN13: 9780198293545
• ISBN10: 0198293542
• Format: Paperback
• Copyright: 2001-01-04
• Publisher: Oxford University Press

Summary

Since the first edition of this book was published in 1993, David Hendry's work on econometric methodology has become increasingly influential. In this edition he presents a brand new paper which compellingly explains the logic of his general approach to econometric modeling and describes recent major advances in computer-automated modeling, which establish the success of the proposed strategy. Empirical studies of consumers' expenditure and money demands illustrate the methods in action. The breakthrough presented here will make econometric testing much easier.

Author Biography

David F. Hendry is Leverhulme Personal Research Professor of Economics and Fellow of Nuffield College, Oxford. He was previously Professor of Econometrics at both the London School of Economics and the University of California at San Diego.

Table of Contents

Preface to the New Edition	xiii
Preface	xv
Acknowledgements	xvii
Introduction	1	(8)
Part I Roots and Route Maps	9	(104)
Econometrics - Alchemy or Science?	11	(18)
Alchemy and Science	11	(1)
Econometrics	12	(2)
Econometrics as Alchemy	14	(6)
Econometrics' Problems	20	(3)
A Structure for Econometrics	23	(4)
Is Econometrics Alchemy or Science?	27	(2)
Stochastic Specification in an Aggregate Demand Model of the United Kingdom	29	(23)
Preamble	29	(3)
Introduction	32	(1)
Methodology: Autocorrelation and Simultaneity	33	(1)
Methodology: Autocorrelation and Dynamics	34	(2)
An Aggregate Demand Model for the United Kingdom, 1957-1967	36	(2)
Methods which Neglect Autocorrelation	38	(1)
Methods Which Treat Autocorrelation but Neglect Simultaneity	39	(5)
A `Limited Information' Treatment of Autocorrelation and Simultaneity	44	(1)
Full Information Estimation of Vector Autocorrelation	44	(4)
Conclusion	48	(4)
Testing Dynamic Specification in Small Simultaneous Systems: an Application to a Model of Building Society Behaviour in the United Kingdom	52	(20)
Preamble	52	(2)
Introduction	54	(1)
Building Societies	55	(10)
The Model of O'Herlihy and Spencer	65	(1)
Statistical Testing of Dynamic Specification in Small Simultaneous Systems	66	(5)
Conclusion	71	(1)
Dynamic Specification	72	(41)
Preamble	72	(2)
Introduction	74	(3)
Data Generation Processes	77	(18)
Finite Distributed Lags	95	(1)
Infinite Distributed Lags	96	(6)
Dynamic Specification in Multi-equation Models	102	(11)
Part II The Development of Empirical Modelling Strategies	113	(174)
On the Time-Series Approach to Econometric Model Building	119	(10)
Preamble	119	(2)
Introduction	121	(1)
Autocorrelation Transforms Applied to Non-stationary Data Series	121	(2)
Differencing Economic Time Series	123	(2)
Simultaneous Equations Systems	125	(1)
Conclusion on Granger and Newbold	126	(1)
On Business Cycle Modelling without Pretending to Have Too Much a priori Economic Theory	126	(3)
Serial Correlation as a Convenient Simplification, not a Nuisance: a Comment on a Study of the Demand for Money by the Bank of England	129	(17)
Preamble	129	(3)
Introduction	132	(1)
COMFAC Analysis	133	(5)
Demand for Money Study of Hacche (1974)	138	(6)
Conclusion and Summary	144	(2)
An Empirical Application and Monte Carlo Analysis of Tests of Dynamic Specification	146	(29)
Preamble	146	(1)
Introduction	147	(1)
Testing Procedures	148	(6)
Empirical Application	154	(5)
Monte Carlo Methods	159	(6)
The Finite Sample Properties of the Tests	165	(7)
Summary and Conclusions	172	(3)
Econometric Modelling of the Aggregate Time-Series Relationship between Consumers' Expenditure and Income in the United Kingdom	175	(35)
Preamble	175	(4)
Introduction	179	(3)
The Data	182	(4)
Three Econometric Studies and their Research Methods	186	(4)
A Standardized Framework	190	(4)
On Multicollinearity	194	(1)
Selection of the `Best' Equation	195	(1)
Measurement Errors	196	(1)
A Simple Dynamic Model	197	(7)
Inflation Effects	204	(4)
Summary and Conclusions	208	(2)
Liquidity and Inflation Effects on Consumers' Expenditure	210	(21)
Preamble	210	(1)
Introduction	211	(2)
Integral Correction Mechanisms	213	(5)
Real Income and Inflation	218	(3)
Empirical Evidence for the United Kingdom	221	(8)
Summary and Conclusions	229	(2)
Interpreting Econometric Evidence: The Behaviour of Consumers' Expenditure in the United Kingdom	231	(15)
Preamble	231	(1)
Introduction	232	(3)
A Reappraisal of DHSY and HUS	235	(4)
The Hall Model	239	(3)
Analysis of Results	242	(2)
`Forward-looking' versus `Backward-looking' Behaviour	244	(2)
Predictive Failure and Econometric Modelling in Macroeconomics: the Transactions Demand for Money	246	(24)
Preamble	246	(2)
Introduction	248	(1)
Predictive Failure and Model Mis-specification	249	(3)
`Simple-to-general' Modelling Methods	252	(4)
From the General to the Specific	256	(4)
Feedback Mechanisms in Differenced Models	260	(5)
Summary and Conclusions	265	(1)
An Empirical Illustration	266	(4)
Monetary Economic Myth and Econometric Reality	270	(17)
Preamble	270	(2)
Present Controvery: a Funeral Pyre or the White Heat of Technical Advance?	272	(2)
Construction and Destruction: Pious Hope versus Hard Reality	274	(2)
Evaluation Criteria: or More Than You Ever Wanted to Know about Testing Models	276	(2)
Empirical Illustration A: or Assertion versus Empirical Evidence	278	(1)
Regime Shifts: or Why Econometrics Textbooks Need to be Rewritten	279	(1)
Empirical Illustration B: UK Money Demand Re-revisited	280	(5)
Empirical Illustration C: US Money Demand Explained	285	(1)
Conclusion	285	(2)
Part III Formalization	287	(130)
The Structure of Simultaneous Equations Estimators	291	(23)
Preamble	291	(2)
Introduction	293	(1)
Simultaneous Equations Estimators	293	(3)
Systems Methods	296	(5)
Individual Equation Methods	301	(5)
Single-equation Methods for Autoregressive Errors	306	(6)
Conclusion	312	(2)
AUTOREG: a Computer Program Library for Dynamic Econometric Models with Autoregressive Errors	314	(16)
Preamble	314	(2)
Econometric Background	316	(4)
Estimator Generation	320	(1)
Numerical Optimization	321	(1)
Method Evaluation	322	(2)
Structure of the Library	324	(2)
The Monte Carlo Programs	326	(1)
Program Validation and Development Stage	327	(1)
Computing Costs	328	(1)
Future Developments	328	(2)
Exogeneity	330	(28)
Preamble	330	(2)
Introduction	332	(2)
Definitions	334	(7)
Examples	341	(6)
Application to Dynamic Simultaneous Equations Models	347	(6)
Summary and Conclusions	353	(5)
On the Formulation of Empirical Models in Dynamic Econometrics	358	(29)
Preamble	358	(4)
Introduction	362	(1)
Theory Models and Empirical Models	363	(2)
An Illustration: the Repayment of Mortgage Principal	365	(3)
An Analysis of Empirical-model Concepts	368	(11)
The Empirical Illustration Reconsidered	379	(3)
Dynamic Simulation	382	(3)
Conclusion	385	(2)
The Econometric Analysis of Economic Time Series	387	(30)
Preamble	387	(1)
Introduction	388	(4)
An Econometric Framework	392	(15)
Estimation	407	(4)
Testing	411	(1)
Model Selection	412	(2)
Conclusion	414	(3)
Part IV Retrospect and Prospect	417	(74)
Econometric Modelling: the `Consumption Function' in Retrospect	419	(25)
Preamble	419	(2)
Introduction	421	(3)
Design Criteria	424	(2)
Data Coherency	426	(3)
Valid Conditioning	429	(2)
Parameter Constancy	431	(3)
Data Admissibility	434	(2)
Theory Consistency	436	(3)
Encompassing	439	(2)
Summary and Conclusion	441	(3)
Postscript: the Econometrics of PC-GIVE	444	(23)
Preamble	444	(1)
An Overview	445	(1)
The Model Class	446	(9)
Model Evaluation	455	(1)
An Information Taxonomy	456	(7)
Test Types	463	(1)
Modelling Strategies	464	(1)
Model Estimation	464	(1)
Conclusion	465	(2)
Epilogue: the Success of General-to-Specific Model Selection	467	(24)
Introduction	467	(1)
Potential Criticisms	468	(8)
Methodological Innovations	476	(3)
Improving the Algorithm	479	(4)
Selection Probabilities	483	(1)
Deletion Probabilities	484	(2)
Improved Inference Procedures	486	(1)
Applying PcGets	487	(2)
Conclusion	489	(2)
References	491	(27)
Bibliography	518	(7)
Index	525