Escolar Documentos
Profissional Documentos
Cultura Documentos
J.:
Introduction to
ECONOMETRICS
Christopher Dougherty
online
f f resource
centre
Introduction to
Econometrics
:
,
11-T} i EDflON
Christopher Dougherty
London School of Economics and Political Science
OXTORD
UNIVERSITY PRESS
OXFORD
UNIVERSITY PRHSS
Introduction to Econometrics
'Fhis is a textbook for a year-long undergraduate course in econometrics. It is
intended to fill a need that has been generated by the changing profile of the
typical econometrics student. Econometrics courses often used to be optional for
economics majors, but now they are becoming compulsory. Several factors are
responsible. Perhaps the most important is the recognition that an understanding
of empirical research techniques is not just a desirable but an essential part of the
basic training of an economist, and that courses limited to applied statistics are
inadequate for this purpose. No doubt this has been reinforced by the fact that
graduate-level courses in econometrics have become increasingly ambitious, with
the consequence that substantial exposure to econometrics at an undergraduate
level is now a requirement for admission to the leading graduate schools.
There are also supply-side factors. The wave that has lifted econometrics to
prominence in economics teaching comes on the heels of another that did the same
for mathematics and statistics. Without this prior improvement in quantitative
training, the shift of econometrics to the core of the economics curriculurn would
not have been possible.
As a consequence of this development, students on econometrics courses are
more varied in their capabilities than ever before. No longer are they a self-
selected minority of mathematical high-fliers. The typical student now is a regular
econorrmics major who has taken basic, but not advanced, courses in calculus
and statistics. The democratization of econometrics has created a need for a
broader range of textbooks than before, particularly for the wider audience. The
mathematical elite has for many years been served hy a number of accomplished
texts. The wider audience has been less well served. This new edition continues to
be chiefly addressed to it.
given the constraints imposed by the nature of its target audience and not making
use of linear algebra.
A primary concern has been not to overwhelm the student with information.
This is not a reference work. It is hoped that the student will find the text readable
and that in the course of a year he or she would comfortably be able to traverse its
contents. For the same reason the mathematical demands on the student have been
kept to a minimum. For nearly everyone, there is a limit to the rate at which formal
mathematical analysis can be digested. lf this limit is exceeded, the student spends
much mental energy grappling with the tcchnicalities rather than the substance,
impeding the development of a unified understanding of the subject.
Although its emphasis is on theory, the text is intended to provide substantial
hands-on practical experience in the forro of regression exercises using a
computer. In particular, the Educational Attainment and Wage Equation data
provide opportunities for 60 cross-sectional exercises spread through the first
10 chapters of the text. Students start with a simple model and gradually develop
it into a more sophisticated one as their knowledge of econometric theory grows.
It is hoped that seeing how the specification of their models improves will motvate
students and help sustain their interest. The Demand Functions data set, with its
15 time series exercises, is intended to provide a similar experience in the remaining
chapters. Further data sets have been provided for specialist applications.
1. The most obvious change in this edition is in the notation. The estimator of the
parameter [3 z is now written (3 2 , instead of b 2 . In introductory texts, the use of
the Latin letter as an estimator of the corresponding Greek one used to be coni-
mon. It had the advantage of making the mathematical analysis slightly cleaner
and simpler. Perhaps it was also less intimidating, an important consideration
for those students who find college algebra challenging. But the more elegant
caret-mark notation has clearly become dominant, even at this level, and so
this text has switched to it.
2. There are 40 new exercises, mostly analytical.
3. The main cross-sectional data set used for examples and computer exercises
has been updated from the National Longitudinal Survey of Youth 1979 to
its successor, the NI,SY 1997.
In addition, there has heen an effort to improve the clarity of the exposition. A
few topics have been added and a few have been deleted, but there has been no
chango to the objective that cach chapter should be readable in its entirety as a
straight narrative.
The intention is to provide enough conceptual material to support a two-semester
scquence, with the Review and the first seven chapters used as an Introducttoii
to the classical linear regression model and the remainder as a second semester
for students who are ready to tackle more sophisticated econometric issues. The
ainl remains that of providing a solid intuitive understanding of the material with
enough technical underpinning to prepare the student for further formal study of
econometrics or for the self-study of applications.
As before, the simulations have p een undertaken using Matt,ab. I will be happy
to email copies of the batch files to anyone interested.
Additional resources
The Online Resource Centre
www.oxfordtextbooks.co.tik/orc/dougherty5e/
offers the following resources for instructors and students:
PowerPoint' slideshows that offer a graphical treatment of most of the topics
in the text. Narrative boxes provide an explanation of the sudes.
Links to data sets and maniials.
Instructor's mannals for the text and data sets, detailing the exercises and
their solutions.
A student area that provides answers to the starred exercises in the text and
offers additional exercises.
It is hoped that the provision of these materials will not only be helpful for the
study of econometrics but also make it satisfying and pleasurable.
Christopher Dougherty
Contents
I NTROI)UCTION 1
7 HETEROSKEDAS'l'ICITY 290
12 At1T000RRELATION 445
1. One is to provide you with the practical skills needed to fit models, given suit-
able data, in a relatively straightforward context. This is fairly easy. Generally,
such applications will be models fitted with cross-sectional data.
2. The second is to prornote the development of an understanding of the statisti-
cal properties of these techniques and hence an understanding of why the
techniques work satisfactorily in certain contexts and not in others. This is
much more demanding.
3. The third, building on the second, is to encourage you to develop a strong
intuitive understanding of the material and with it the capacity and confidence
to extend it further, either sideways, in applications in a particular field, or
vertically, moving on to more advanced study.
Additional resources
There are two additional major resources that you should check out as soon as
you begin to use this text: the slideshows and the study guide. Roth are avail-
able, at no cost and with no restrictions, in the Online Resource Centre at
www.oxfordtextbooks.co.uk/orc/doughertySe/.
Slideshows: The PowerPoint slideshows systematicaily cover all of the topics
treated in the text, typically with greater graphical detail. They are not intended
as a substituta for the text, but they should provide substantial support.
Study guide: This provides answers to the starred exercises in the text and addi-
tional exercises, also with solutions. It was commissioned by the University of
London International Programmes as an additional resource for distance-learning
students, and the organizers of the External Negree have kindly allowed it to be
available to anyone who is interested in using it.
The Online Resource Centre also gives unrestricted access to ah l of the data sets
used in the examples and exercises in the text.
4 Introduction
Econometrics software
There are at least ten major commercial software packages for econometrics in
use around the world and it does not matter which one you use. With little vari-
ation, they al! have the features and facilities used in econometrics at this leve!.
Many of the tables in this text reproduce output from Stata or EViews, mainly
because the format is compact and tidy. Output from other applications looks
very similar.
If you do not have access to one of these -commercial applications, then down-
load gretl and use that instead. gretl is a powerful, sophisticated econometrics
application, which is easy to use and free. Go to the Online Resource Centre, find
the link, and follow the instructions. There you will also find a downloadable
manual that tells you how to use gretl to do the exercises in this tcxt.
You should not try to use an inferior substitute. In particular, you should not try
to use the regression engine built into a spreadsheet application such as Microsoft
Excel. Excel and other spreadsheets are invaluable applications, but they are not
intended or designed for serious econometrics use. You need a dedicated applica-
tion, and gretl is an excellent one.
The aims of this text have been stated abo y e. There is one further aim, or at
least, hope. That is that you will find the study of econometrics intellectually
satisfying. By the time that you approach the end of this text, you will find that,
although the material in each chapter is new, the sane themes and concerns keep
reappearing, especially those related to the properties of estimators. When you
begin to recognize this, you will be well on your way to becoming a proper econo-
metrician, and not just someone mechanically handling data and performing tests.
And, of course, when the time comes for you to fit your own modeis with your
own data, it is hoped that you will find the practice of econometrics enjoyable too.
Amemiya, Takeshi (1981). Qualitative response models: a survey. Journal of Economic
Literature 19(4): 1483-1536.
Amemiya, Takeshi (1984). Tobit models: a survey. Journal of Econometrics 24(1): 3-61.
Baltagi, Badi H. (2013). Econometric Analysis of Panel Data (5th edn). Chichester:
Wiley.
Box, George E.P., and David R. Cox (1964). An analysis of transformations. Journal of the
Royal Statistical Society Series B 26(2): 211-43.
Box, George E.P., and Norman R. Draper (1987). Empirical Model-Building and Response
JKIfC!(.J. ivcw 1UlK. w L1 y.
Box, George E.P., and Gwilym M. Jenkins (1970). Time Series Analysis: Forecasting and
Control. San Francisco: Holden Day.
Box, George E.P., Gwilym M. Jenkins, and Gregory C. Reinsel (1994). Time Series Analysis:
Forecasting and Control (3rd edn). Englewood Cliffs, NJ: Prentice-Hall.
Breusch, Trevor S. (1978). Testing for autocorrelation in dynamic linear models. Australian
Economic Papers 17(31): 334-55.
Brown, T.M. (1952). Habit persistence and lags in consumer behaviour. Econometrica
20(3): 355-71.
Card, David (1995). Using geographic variation in college proximity to estimate the return
to schooling. In Louis N. Christofides, E. Kenneth Grant, and Robert Swidinsky (eds),
Aspects of Labour Market Behaviour: Essays in Honour of John Vanderkamp. Toronto:
University of Toronto Press.
Chow, Gregory C. (1960). Tests of equality between sets of coefficients in two linear regres-
sions. Econometrica 28(3): 591-605.
Cobb, Charles W., and Paul H. Douglas (1928). A theory of production. American
Economic Review 18(1, Suppl.): 139-65.
Cooper, Ronald L. (1972). The predictive performance of quarterly econometric models of
the United States. In Bert G. Hickman (ed.), Econometric Models of Cyclical Behavior,
Vol. II. New York: Columbia University Press.
Court, Andrew T. (1939). Hedonic price indexes with automotive examples. In The
Dynamics of Automobile Demand, Papers presented at a joint meeting of the American
Statistical Association and the Econometric Society in Detroit, December 1938. New
York: General Motors Corporation.
Davidson, James E.H. (2000). Econometric Theory. Oxford: Blackwell.
Davidson, Russell, and James G. MacKinnon (1993). Estimation and Inference in
Econometrics. New York: Oxford University Press.
578 Bibliography
Dickey, David A., and Wayne A. Fuller (1979). Distribution of the estimators for autore-
gressive time series with a unit root. Journal of the American Statistical Association
74(366): 427-31.
Dickey, David A., and Wayne A. Fuller (1981). Likelihood ratio statistics for autoregressive
time series with a unit root. Econometrica 49(4): 1057-72.
Diebold, Francis X. (1998). The past, present, and future of macroeconomic forecasting.
Journal of Economic Perspectives 12(2): 175-92.
Diebold, Francis X. (2001). Elements of Forecasting (2nd edn). Cincinnati, OH:
South-Western.
Durbin, James (1954). Errors in variables. Review of the International Statistical Institute
22(1): 23-32.
Durbin, James (1970). Testing for serial correlation in least-squares regression when some
of the regressors are lagged dependent variables. Econometrica 38(3): 410-21.
Durbin, James, and G.S. Watson (1950). Testing for serial correlation in least-squares
regression I. Biometrika 37(3-4): 409-28.
Durlauf, Steven N., and Peter C.B. Phillips (1988). Trends versus random walks in time
series analysis. Econometrica 56(6): 1333-54.
Elliott, Graham, Thomas J. Rothenberg, and James H. Stock (1996). Efficient tests for an
autoregressive unit root. Econometrica 64(4): 813-36.
Engle, Robert F., and Clive W.J. Granger (1987). Co-integration and error correction rep-
resentation, estimation, and testing. Econometrica 50(2): 251-76.
Engle, Robert F., and Clive W.J. Granger (1991). Long-Run Economic Relationships:
Readings in Cointegration (editors). Oxford: Oxford University Press.
Fowler, Floyd J. (2009). Survey Research Methods (4th edn). Thousand Oaks, CA: Sage.
Friedman, Milton (1957). A Theory of the Consumption Function. Princeton, Ni: Princeton
University Press.
Frisch, Ragnar, and Frederick V. Waugh (1933). Partial time regressions as compared with
individual trends. Econometrica 1(4): 387-401.
Godfrey, Leslie G. (1978). Testing against general autoregressive and moving average error
models when the regressors include lagged dependent variables. Econometrica 46(6):
1293-1301.
Goldfeld, Stephen M., and Richard E. Quandt (1965). Some tests for homoscedasticity.
Journal of the American Statistical Association 60(310): 539-47.
Granger, Clive W.J., and Paul Newbold (1974). Spurious regressions in econometrics.
Journal of Econometrics 2(2): 111-20.
Greene, William (2011). Econometric Analysis (7th edn). Upper Saddle River, NJ: Prentice
Hall.
Gronau, Reuben (1974). Wage comparisons-a selectivity bias. Journal of Political
Economy 82(6): 1119-55.
Hamilton, James D. (1994). Time Series Analysis. Princeton, NJ: Princeton University Press.
Hausman, Jerry A. (1978). Specification tests in econometrics. Econometrica 46(6):
1251-71.
Heckman, James (1976). The common structure of statistical models of truncation, sam-
ple selection, and limited dependent variables and a simple estimator for such models.
Annals of Economic and Social Measurement 5(4): 475-92.
Hendry, David E (1979). Predictive failure and econometric modelling in macroeconomics:
the transactions demand for money. In Paul Ormerod (ed.), Modelling the Economy.
London: Heinemann.
Introduction to Econometrics 579
Hendry, David E, and Grayham E. Mizon (1978). Serial correlation as a convenient simpli-
fication, not a nuisance. Economic Journal 88(351): 549-63.
Holden, Darryl, and Roger Perman (2007). Unit roots and cointegration for the econo-
mist. In B. Bhaskara Rao (ed.), Cointegration for the Applied Economist (2nd edn).
Basingstoke: Paigrave Macmillan.
Hsiao, Cheng (2015). Analysis of Panel Data (3rd edn). Cambridge: Cambridge University
Press.
Kalecki, Micha! (1935). A macrodynamic theory of business cycles. Econometrica 3(3):
327-44.
Koyck, Leendert M. (1954). Distributed Lags and Investment Analysis. Amsterdam:
North-Holland.
Lintner, John (1956). Distribution of incomes of corporations among dividends, retained
earnings, and taxes. American Economic Review 46(2): 97-113.
Liviatan, Nissan (1963). Tests of the Permanent-Income Hypothesis based on a reinterview
savings survey. In Carl Christ (ed.), Measurement in Economcs. Stanford, CA: Stanford
University Press.
Lovell, Michael C. (1963). Seasonal adjustment of economic time series. Journal of the
American Statistical Association 58: 993-1010.
MacKinnon, James G., and Halbert White (1985). Some heteroskedasticity-consistent
covariance matrix estimators with improved finite sample properties. Journal of
Econometrics 29(3): 305-25.
Nelson, Charles R., and Charles I. Plosser (1982). Trends and random walks in macro-
economic time series: some evidence and implications. Journal of Monetary Economics
10(2): 139-62.
Nerlove, Marc (1963). Returns to scale in electricity supply. In Carl Christ (ed.),
Measurement in Economics. Stanford, CA: Stanford University Press.
Park, Rolla E., and Bridget M. Mitchell (1980). Estimating the autocorrelated error model
with trended data. Journal of Econometrics 13(2): 185-201.
Peach, James T., and James L. Webb (1983). Randomly specified macroeconomic models:
some implications for model selection. Journal of Economic Issues 17(3): 697-720.
Phillips, Peter C.B. (1986). Understanding spurious regressions in econometrics. Journal of
Econometrics 33(3): 311-40.
Rubin, Herman (1950). Consistency of maximum-likelihood estimates in the explosive
case. In T.C. Koopmans (ed.), Statistical Inference in Dynamic Economic Models. New
York: John Wiley.
Salkever, David S. (1976). The use of dummy variables to compute predictions, prediction
errors and confidence intervals. Journal of Econometrics 4(4): 393-7.
Sims, Christopher A. (1980). Macroeconomics and reality. American Economic Review
48(1): 1-48.
Stock, James H. (1987). Asymptotic properties of least squares estimators of cointegrating
vectors. Econometrica 55(5): 1035-56.
Tinbergen, Jan (1939). Statistical Testing of Business Cycle Theories. No. 2, Business
Cycles in the United States of America 1919-1932. Geneva: League of Nations.
Tobin, James (1958). Estimation of relationships for limited dependent variables.
Econometrica 26(1): 24-36.
Waugh, Frederick V. (1929). Quality as a Determinant of Vegetable Prices. New York:
Columbia University Press.
580 Bibliography
Random effects regression 537-9; see also Panel unbiasedness 122, 313-14
data two explanatory variables
Random variables analytical decomposition 165
continuous 7, 14-19 derivation of expressions 158-60
discrete 7-13 population variance 166-7
double structure 24 standard errors 167-9
expected value 8-9, 18-19 unbiasedness 165-6
fixed and random components 12-13 Regression model
independence of two random variables 19 assumptions 114-18, 164-5, 311-13, 405-8
standard deviation 11 fitted 87
variance 11-12, 18-19 Model A 113
Random walk 485 Model B 114, 311
with drift 487 Model C 114,405
Granger-Newbold spurious regressions Regressor (explanatory variable, independent
Rank condition for identification 359 variable)
Realization 23, 479; see also Data generation nonstochastic 118, 122
process stochastic 311
Reduced form equation 344; see also reparameterizaton of model specification
Simultaneous equations estimation estimation of long-run effects in dynamic
Redundant variable see Model misspecification model 415-16, 418
Reference category see Dummy variables standard error of linear combination of
Regression analysis, ordinary least squares parameters 282-3
(OLS) t test of linear restriction 282-5
simple regression analysis 85-89 RESET test 222
least squares criterion 88, 91, 94 Residual
multiple regression analysis 156-62 definition of 88
normal equations 93, 159-61 OLS regressions with intercept
see also Disturbance term; Nonlinear zero correlation with explanatory
regression analysis; R 2 ; Regression variables 107-8
model assumptions; Residual zero sample mean 106
Regression coefficients, IV see Instrumental use of outliers in improving model
variables specification 269
Regression coefficients, OLS Residual sum of squares (RSS) 88
as random variables 122-4 Restriction
asymptotic properties, Model B 315-17 benefits from exploitation 179, 280
asymptotic normality 317 definition
consistency 316 linear restriction 179, 280
reason for interest 315 nonlinear restriction 456
confidente intervals 147-8, 169 tests
effects of changes in units of variables 100-1 common factor test of nonlinear
hypothesis testing 139-44, 169; see also t tests restriction 460-2
inconsistency caused by likelihood ratio test 400
measurement error in explanatory F test of linear restriction 281-2
variable 318-20 F test of multiple linear restrictions 285
simultaneous equations bias 345-7 t test of linear restriction 284-5
interpretation use in mitigation of problem of
logarithmic model 202 multicollinearity 179, 280
multiple linear regression model 161-3 zero restrictions 285-6
semilogarithmic model 206 RSS see Residual sum of squares
simple linear regression model 98-100
one explanatory variable Sample selection model 386-90
analytical decomposition 118-22 Heckman two-step estimation procedure 387,
consistency 316 388
derivation of expressions 92-4 Sample mean see Mean of a random variable
Monte Carlo experiment 126-30 Sample selection bias see Sample selection model
population variante 130-3 School costs (SC) data set 573-4
standard errors 133-6 Schwarz Information Criterion (SIC) 510
Subject Index 59
The most accessible econometrics text focusing on only the essential maths.
Keeping maths to a minimum, this book provides a non-technical introduction to econometrics, making
it the perfect companion for anyone new to the subject.
A revision chapter at the beginnng of the book gives you the opportunity to brush up on statistics,
whilst diagrams have been included wherever possible to ensure clarity of explanation. Packed
with plenty of examples and regression exercises, including 50 on the same data set, Introduction to
Econometrics gives you lots of hands-on experience and ensures that you hone the skills needed to
successfully fit models given suitable data.
Whatever your level of experience, this book will develop your confidence in econometrics, providing
a launch pad for further study and equipping you with the tools to answer economic questions.
a)
PowerPoint sudes covering all the topics in the text Instructor's manual containing answers to the
Datasets from the text available in Stata, Excel, exercises in the text a
Eviews, and ASCII formats Instructor PowerPointslides
Study guide with further exercises Ei
a),
>;
I
ISBN 978-0-19-967682-8
OXFORD
UNIVERSITY PRESS
9 780199 676828
www.ouv.com