Dynamic Econometrics

984 Pages·1995·59.376 MB·English

by  David F. Hendry

Title Pages
Dynamic Econometrics
David F. Hendry
Print publication date: 1995
Print ISBN-13: 9780198283164
Published to Oxford Scholarship Online: November 2003
DOI: 10.1093/0198283164.001.0001
Title Pages
(p.i) Advanced Texts in Econometrics
(p.iii) Dynamic Econometrics
(p.iv)
Great Clarendon Street, Oxford OX2 6DP
Oxford University Press is a department of the University of Oxford
It furthers the University's objective of excellence in research,
scholarship,
and education by publishing worldwide in
Oxford New York
Auckland Bangkok Buenos Aires Cape Town Chennai
Dar es Salaam Delhi Hong Kong Istanbul Karachi Kolkata
Kuala Lumpur Madrid Melbourne Mexico City Mumbai Nairobi
São Paulo Shanghai Taipei Tokyo Toronto
Oxford is a registered trade mark of Oxford University Press
in the UK and in certain other countries
Published in the United States by
Oxford University Press Inc., New York
© David F. Hendry 1995
The moral rights of the authors have been asserted
Database right Oxford University Press (maker)
Page 1 of 2
Title Pages
First published 1995
All rights reserved. No part of this publication may be reproduced,
stored in a retrieval system, or transmitted, in any form or by any
means,
without the prior permission in writing of Oxford University Press,
or as expressly permitted by law, or under terms agreed with the
appropriate
reprographics rights organization. Enquiries concerning
reproduction
outside the scope of the above should be sent to the Rights
Department,
Oxford University Press, at the address above
You must not circulate this book in any other binding or cover
and you must impose this same condition on any acquirer
British Library Cataloguing in Publication Data
Data available
Library of Congress Cataloging in Publication Data
Data available
ISBN 0‐19‐828317‐2 (Hbk)
ISBN 0‐19‐828316‐4 (Pbk)
Page 2 of 2
Dedication
Dynamic Econometrics
David F. Hendry
Print publication date: 1995
Print ISBN-13: 9780198283164
Published to Oxford Scholarship Online: November 2003
DOI: 10.1093/0198283164.001.0001
Dedication
(p.v) To Evelyn and Vivien (p.vi)
Page 1 of 1
Epigraph
Dynamic Econometrics
David F. Hendry
Print publication date: 1995
Print ISBN-13: 9780198283164
Published to Oxford Scholarship Online: November 2003
DOI: 10.1093/0198283164.001.0001
Epigraph
(p.vii) ‘The only way to a position in which our science might give positive
advice on a large scale to politicians and business men, leads through
quantitative work. For as long as we are unable to put our arguments into
figures, the voice of our science, although occasionally it may help to dispel
gross errors, will never be heard by practical men. They are, by instinct,
econometricians all of them, in their distrust of anything not amenable to exact
proof.’
From Joseph A. Schumpeter: ‘The Common Sense of
Econometrics’, Econometrica, 1 (1933), p. 12. (p.viii)
Page 1 of 1
Figures
Dynamic Econometrics
David F. Hendry
Print publication date: 1995
Print ISBN-13: 9780198283164
Published to Oxford Scholarship Online: November 2003
DOI: 10.1093/0198283164.001.0001
(p.xxiii) Figures
1.1 Histograms of economic time series 15
1.2 UK, US and artificial time series 20
1.3 Time series of the UK price level and its differences 21
1.4 UK money growth in different sub‐samples 27
2.1 UK inflation 32
2.2 Histograms and densities for aggregate price data 35
2.3 Histograms and densities for artificial data 36
2.4 Empirical correlograms 41
2.5 Artificial and economic time series 43
2.6 Correlograms after removing lagged information 60
3.1 Graphical regression statistics 77
3.2 Graphical recursive statistics 80
3.3 Graphical recursive evaluation statistics 81
3.4 Histogram and density for the autoregressive coefficient, α=0.5 96
3.5 Simulated N[0, 1] 96
3.6 Recursive Monte Carlo statistics 98
3.7 Hypothetical distributions for UK GNP 99
3.8 Constructing the step function 102
3.9 Random walks over increasing sample sizes 103
3.10 Histogram of χ2(1) 110
3.11 Recursive Monte Carlo outcomes 116
3.12 Histogram and density of β̂ when β = 1 117
3.13 Histogram and density of β̂ when β = 0.9 117
3.14 Histogram and density of β̂ when β = 0.5 117
3.15 Finite sample and asymptotic standard errors for β = 0.9, 0.5 118
4.1 Frequency distribution of β̂ 124
1
4.2 Frequency distribution of the t‐test of H : β = 0 125
0 1
4.3 Recursive outcomes for the nonsense regression simulation 126
Page 1 of 4
Figures
4.4 Frequency distribution of R for two unrelated I(0) time series 127
4.5 Frequency distribution of R for two unrelated I(1) time series 128
4.6 Frequency distribution of R for two unrelated I(2) time series 128
4.7 Frequency distribution of DW for two unrelated I(1) time series 130
4.8 Monte Carlo rejections for different values of α 147
(p.xxiv)
4.9 Recursive bias estimates 150
5.1 Parameter space restrictions 163
5.2 Standardized frequency distribution of OLS biases in case (d) 189
5.3 Recursively computed biases and rejection frequencies 190
6.1 Linear approximations to a non‐linear relation 198
6.2 Predictors based on data differences 206
6.3 Recursive estimation statistics 208
6.4 Disturbance correlograms and densities 209
6.5 Plot of RSS (α ) for (6.33) 211
1
6.6 Normalized and cumulative lag weights 217
6.7 Levels and differences of artificial data 219
6.8 Recursive Monte Carlo output 221
6.9 Frequency distribution of β̂ 222
2
6.10 Frequency distribution of β̂ 222
1
6.11 Time series of velocity and the net interest rate 224
6.12 UK M1 regression fits, forecasts and residuals 225
6.13 Parameter constancy statistics 226
7.1 Fitted values and outcomes from the static regression for M1 236
7.2 Recursive statistics from the static regression for M1 237
7.3 Histogram and density function for γ̂ 238
7.4 Histogram and density function for the t‐test of H : γ = 0 239
0
7.5 Recursive Monte Carlo outcomes for the static regression 240
7.6 Fitted values and outcomes from the autoregression for M1 244
7.7 Recursive statistics from the autoregression for M1 245
7.8 Histogram and density for ρ̂ 246
7.9 Histogram and density for the t‐test of H : ρ = 0 246
0
7.10 Recursive Monte Carlo statistics for the autoregression 247
7.11 Fitted and actual values for the growth‐rate model of M1 249
7.12 Recursive statistics for the growth‐rate model of M1 250
7.13 Histogram and density for φ 251
7.14 Recursive Monte Carlo outcomes for the growth‐rate model 252
7.15 Fitted and actual values for the leading indicator model of M1 253
7.16 Recursive statistics for the leading indicator model of M1 254
7.17 Histogram and density for the leading indicator model 255
7.18 Recursive Monte Carlo outcomes for the leading indicator model 256
7.19 Fitted and actual values from the partial adjustment model for M1
263
7.20 Recursive statistics from the partial adjustment model for M1 264
Page 2 of 4
Figures
7.21 Histogram and density for β̂ in the Monte Carlo for partial
1
adjustment 264
7.22 Histogram and density for β̂ in the Monte Carlo for partial
2
adjustment 265
7.23 Histogram and density for the t‐test of H : b = β 265
0 2 2
7.24 Recursive Monte Carlo statistics for partial adjustment 266
7.25 Fitted values and outcomes for the COMFAC model of M1 271
(p.xxv)
7.26 RSS as a function of the autocorrelation parameter 272
7.27 Fitted and actual values for the distributed‐lag model of M1 279
7.28 Recursive statistics for the distributed‐lag model of M1 280
7.29 Histogram and density function for ω̂ 281
1
7.30 Histogram and density function for the DW test 281
7.31 Recursive Monte Carlo statistics for the distributed‐lag model 282
7.32 Fitted and actual values for the dead‐start model 284
7.33 Recursive statistics for the dead‐start model 285
7.34 Histogram and density function for β̂ 285
2
7.35 Recursive Monte Carlo statistics for the dead‐start model 286
7.36 Fitted and actual values from the equilibrium‐correction model of
M1 292
7.37 Recursive statistics for the equilibrium‐correction model of M1 293
7.38 Histogram and density function for (β̂ − 1) 293
2
7.39 Recursive Monte Carlo statistics for the equilibrium‐correction
model 294
7.40 Lag weights in the Monte Carlo study of the model typology 305
8.1 Fitted and actual values and residuals 325
8.2 Residual correlograms and densities 326
8.3 Cointegrating vector time series 327
8.4 Recursive outcomes 328
8.5 Fitted and actual values and residuals for the parsimonious system
336
8.6 Dynamic simulation of closed and open I(1) and I(0) systems 342
10.1 Density and likelihood functions 376
10.2 Score equation 382
11.1 Restricted cointegration vector and recursive eigenvalues 420
11.2 Log‐likelihood grids 437
11.3 Model recursive estimates 438
11.4 12‐quarter ahead forecasts 439
12.1 Price index revisions to the US GNP deflator 462
12.2 Monte Carlo recursive biases from measurement errors 465
14.1 Recursive graphical statistics for Δ R 537
14.2 Recursive graphical statistics for Δ log M/PY 538
15.1 Seasonal behaviour of UK consumption and money stock 560
15.2 UK consumption and income 561
Page 3 of 4
Figures
15.3 Functions of monthly US long‐term interest rates 569
16.1 UK and US money stocks, output, and prices 579
16.2 Weighting function for learning adjustment of interest rates 585
16.3 Interest rates, real money and expenditure growth, and inflation 586
16.4 Money growth, inflation, velocity and interest rates 587
(p.xxvi)
16.5 Data densities for money, inflation, income and interest rates 588
16.6 Correlograms for money, inflation, income and interest rates 589
16.7 Fitted and actual values for trend equation 590
16.8 System fitted and actual values, and residuals 594
16.9 System recursive evaluation statistics 595
16.10 Graphical diagnostic information 596
16.11 β̂′ x and recursive eigenvalues 598
t
16.12 Fitted values, outcomes and residuals for the model 601
16.13 Recursive FIML statistics 602
16.14 One‐dimensional projections of the likelihood surface 603
16.15 One‐step model‐based forecasts 605
16.16 Dynamic model‐based forecasts 606
16.17 Graphical evaluation statistics for M1‐demand GUM 610
16.18 Graphical evaluation statistics for M1‐demand equation 613
16.19 Recursive OLS statistics for M1‐demand equation 614
16.20 Recursive statistics for the inflation equation 617
A2.1 Four empirical histograms with smoothed approximating shapes 640
A2.2 Set relationships 644
A2.3 Probability relations 651
A2.4 Distributional shapes 673
A4.1 Central‐limit convergence 716
A4.2 Convergence of OLS in a dynamic equation 727
A4.3 Behaviour of IV estimation in a just‐identified equation 729
A5.1 Climbing a one‐dimensional hill 753
A5.2 Fitting a quadratic in a line search 754
A5.3 Climbing three‐dimensional hills 755
A5.4 Autoregressive error function grid 762
A5.5 Gradient optimization 766
A6.1 UK quarterly macroeconomic time series 797
A6.2 Further UK quarterly macroeconomic time series 804
A6.3 Time series and fits for C, Y, Δ P, and Q 812
A6.4 Dynamic forecast of C, Y, Δ P, and Q 813
A6.5 Dynamic forecast of C, Y, Δ P, and Q with error bars 815
(p.xxvii)
Page 4 of 4
Tables
Dynamic Econometrics
David F. Hendry
Print publication date: 1995
Print ISBN-13: 9780198283164
Published to Oxford Scholarship Online: November 2003
DOI: 10.1093/0198283164.001.0001
Tables
1.1 Sub‐sample estimates 26
3.1 Correlation structure 77
3.2 Convergence results for normalized sample moments 107
3.3 Monte Carlo outcomes 115
4.1 Monte Carlo outcomes for nonsense regressions 129
4.2 Monte Carlo MCSD and ESE 149
7.1 Autoregressive‐distributed lag typology 232
7.2 AD(0,3) mean coefficient estimates 280
8.1 Vector autoregressive‐distributed lag system typology 322
8.2 Summary statistics for the VAR 326
8.3 Summary statistics for VECM 336
8.4 Summary statistics for (8.70)–(8.71) 337
9.1 Evaluation and design criteria 366
11.1 Summary statistics 438
12.1 ECM measurement error biases 465
13.1 Hypothesis forms 486
13.2 Significance levels for t(50) 491
14.1 Summary statistics 526
14.2 Diagnostic test outcomes 526
14.3 Encompassing test statistics 527
14.4 Revised encompassing test statistics 527
15.1 Unrestricted money‐demand equation 549
15.2 Restricted money‐demand equation 549
15.3 Four‐variable cointegration analysis 551
15.4 Three‐variable cointegration analysis 552
16.1 Residual correlations 593
16.2 Unrestricted VAR estimates 593
(p.xxviii)
Page 1 of 2