(eBook PDF) Econometrics by Example 2nd Edition

instant download

https://ebooksecure.com/product/ebook-pdf-econometrics-by-
example-2nd-edition/

Download more ebook from https://ebooksecure.com

We believe these products will be a great fit for you. Click
the link to download now, or visit ebooksecure.com
to discover even more!

Your Wish Is My Command: Programming by Example - eBook

PDF

https://ebooksecure.com/download/your-wish-is-my-command-
programming-by-example-ebook-pdf/

(eBook PDF) Introductory Econometrics: Asia-Pacific 2nd

Edition

http://ebooksecure.com/product/ebook-pdf-introductory-
econometrics-asia-pacific-2nd-edition/

(eBook PDF) Practical Econometrics by Hilmer

http://ebooksecure.com/product/ebook-pdf-practical-econometrics-
by-hilmer/

(eBook PDF) Essential Statistics, Regression, and

Econometrics 2nd Edition

http://ebooksecure.com/product/ebook-pdf-essential-statistics-
regression-and-econometrics-2nd-edition/
(eBook PDF) Essential Statistics, Regression, and
Econometrics 2nd Edition

http://ebooksecure.com/product/ebook-pdf-essential-statistics-
regression-and-econometrics-2nd-edition-2/

Mathematica by Example 5th Edition Martha L. Abell &

James P. Braselton - eBook PDF

https://ebooksecure.com/download/mathematica-by-example-ebook-
pdf/

(eBook PDF) The Econometrics of Financial Markets 2nd

ed. Edition

http://ebooksecure.com/product/ebook-pdf-the-econometrics-of-
financial-markets-2nd-ed-edition/

(eBook PDF) Basic Econometrics 5th Edition by Gujarati

http://ebooksecure.com/product/ebook-pdf-basic-econometrics-5th-
edition-by-gujarati/

Progress in Heterocyclic Chemistry Volume 29 1st

Edition - eBook PDF

https://ebooksecure.com/download/progress-in-heterocyclic-
chemistry-ebook-pdf/
 CONTENTS

Preface xiii
Acknowledgments xvii
A personal message from the author xx
List of tables xxii
List of figures xxix

Part I: Basics of linear regression

1 The linear regression model: an overview 2

1.1 The linear regression model 2
1.2 The nature and sources of data 5
1.3 Estimation of the linear regression model 6
1.4 The classical linear regression model (CLRM) 8
1.5 Variances and standard errors of OLS estimators 10
1.6 Testing hypotheses about the true or population regression coefficients 11
1.7 R2: a measure of goodness of fit of the estimated regression 13
1.8 An illustrative example: the determinants of hourly wages 14
1.9 Forecasting 19
1.10 The road ahead 19
Exercises 22
Appendix: The method of maximum likelihood (ML) 25
2 Functional forms of regression models 28
2.1 Log-linear, double log or constant elasticity models 28
2.2 Testing validity of linear restrictions 32
2.3 Log-lin or growth models 33
2.4 Lin-log models 36
2.5 Reciprocal models 38
2.6 Polynomial regression models 40
2.7 Choice of the functional form 42
2.8 Comparing linear and log-linear models 43
2.9 Regression on standardized variables 44
2.10 Regression through the origin: the zero-intercept model 46
2.11 Measures of goodness of fit 49
2.12 Summary and conclusions 50
VIII CONTENTS

Exercises 51
3 Qualitative explanatory variables regression models 53
3.1 Wage function revisited 53
3.2 Refinement of the wage function 55
3.3 Another refinement of the wage function 56
3.4 Functional form of the wage regression 59
3.5 Use of dummy variables in structural change 61
3.6 Use of dummy variables in seasonal data 64
3.7 Expanded sales function 66
3.8 Piecewise linear regression 69
3.9 Summary and conclusions 73
Exercises 74

Part II: Regression diagnostics

4 Regression diagnostic I: multicollinearity 80

4.1 Consequences of imperfect collinearity 81
4.2 An example: married women’s hours of work in the labor market 84
4.3 Detection of multicollinearity 85
4.4 Remedial measures 87
4.5 The method of principal components (PC) 89
4.6 Summary and conclusions 92
Exercises 93
5 Regression diagnostic II: heteroscedasticity 96
5.1 Consequences of heteroscedasticity 96
5.2 Abortion rates in the USA 97
5.3 Detection of heteroscedasticity 100
5.4 Remedial measures 103
5.5 Summary and conclusions 110
Exercises 110
6 Regression diagnostic III: autocorrelation 113
6.1 US consumption function, 1947–2000 113
6.2 Tests of autocorrelation 115
6.3 Remedial measures 121
6.4 Model evaluation 126
6.5 Summary and conclusions 129
Exercises 129
7 Regression diagnostic IV: model specification errors 131
7.1 Omission of relevant variables 131
7.2 Tests of omitted variables 135
7.3 Inclusion of irrelevant or unnecessary variables 138
7.4 Misspecification of the functional form of a regression model 139
7.5 Errors of measurement 141
7.6 Outliers, leverage and influence data 142
7.7 Probability distribution of the error term 145
7.8 Random or stochastic regressors 147
 CONTENTS IX

7.9 The simultaneity problem 147

7.10 Dynamic regression models 153
7.11 Summary and conclusions 162
Exercises 163
Appendix: Inconsistency of the OLS estimators of the
consumption function 167 I
Part III: Topics in cross-section data

8 The logit and probit models 170

8.1 An illustrative example: to smoke or not to smoke 170
8.2 The linear probability model (LPM) 171
8.3 The logit model 172
8.4 The language of the odds ratio (OR) 180
8.5 The probit model 181
8.6 Summary and conclusions 184
Exercises 185
9 Multinomial regression models 190
9.1 The nature of multinomial regression models 190
9.2 Multinomial logit model (MLM): school choice 192
9.3 Conditional logit model (CLM) 198
9.4 Mixed logit (MXL) 201
9.5 Summary and conclusions 201
Exercises 203
10 Ordinal regression models 206
10.1 Ordered multinomial models (OMM) 207
10.2 Estimation of ordered logit model (OLM) 207
10.3 An illustrative example: attitudes toward working mothers 209
10.4 Limitation of the proportional odds model 212
10.5 Summary and conclusions 215
Exercises 216
Appendix: Derivation of Eq. (10.4) 218
11 Limited dependent variable regression models 219
11.1 Censored regression models 220
11.2 Maximum likelihood (ML) estimation of the censored regression
model: the Tobit model 223
11.3 Truncated sample regression models 227
11.4 A concluding example 229
11.5 Summary and conclusions 232
Exercises 233
Appendix: Heckman’s (Heckit) selection-bias model 234
12 Modeling count data: the Poisson and negative binomial regression models 236
12.1 An illustrative example 236
12.2 The Poisson regression model (PRM) 238
12.3 Limitation of the Poisson regression model 242
12.4 The Negative Binomial Regression Model (NBRM) 244
X CONTENTS

12.5 Summary and conclusions 244

Exercises 245

Part IV: Time series econometrics

13 Stationary and nonstationary time series 250

13.1 Are exchange rates stationary? 250
13.2 The importance of stationary time series 251
13.3 Tests of stationarity 251
13.4 The unit root test of stationarity 255
13.5 Trend stationary vs. difference stationary time series 258
13.6 The random walk model (RWM) 262
13.7 Summary and conclusions 266
Exercises 267
14 Cointegration and error correction models 269
14.1 The phenomenon of spurious regression 269
14.2 Simulation of spurious regression 270
14.3 Is the regression of consumption expenditure on disposable income
spurious? 271
14.4 When a spurious regression may not be spurious 274
14.5 Tests of cointegration 275
14.6 Cointegration and error correction mechanism (ECM) 276
14.7 Are 3-month and 6-month Treasury Bill rates cointegrated? 278
14.8 Summary and conclusions 280
Exercises 281
15 Asset price volatility: the ARCH and GARCH models 283
15.1 The ARCH model 284
15.2 The GARCH model 290
15.3 Further extensions of the ARCH model 292
15.4 Summary and conclusions 294
Exercises 295
16 Economic forecasting 296
16.1 Forecasting with regression models 296
16.2 The Box–Jenkins methodology: ARIMA modeling 302
16.3 An ARMA model of IBM daily closing prices, 3 January 2000 to
31 October 2002 304
16.4 Vector autoregression (VAR) 310
16.5 Testing causality using VAR: the Granger causality test 315
16.6 Summary and conclusions 319
Exercises 320
Appendix: Measures of forecast accuracy 323

Part V: Selected topics in econometrics

17 Panel data regression models 326

17.1 The importance of panel data 326
17.2 An illustrative example: charitable giving 327
 CONTENTS XI

17.3 Pooled OLS regression of charity function 328

17.4 The fixed effects least squares dummy variable (LSDV) model 330
17.5 Limitations of the fixed effects LSDV model 332
17.6 The fixed effect within group (WG) estimator 333
17.7 The random effects model (REM) or error components model (ECM) 335
17.8 Fixed effects model vs. random effects model 336 I
17.9 Properties of various estimators 339
17.10 Panel data regressions: some concluding comments 339
17.11 Summary and conclusions 340
Exercises 341
18 Survival analysis 344
18.1 An illustrative example: modeling recidivism duration 344
18.2 Terminology of survival analysis 345
18.3 Modeling recidivism duration 348
18.4 Exponential probability distribution 348
18.5 Weibull probability distribution 351
18.6 The proportional hazard model 353
18.7 Summary and conclusions 355
Exercises 356
19 Stochastic regressors and the method of instrumental variables 358
19.1 The problem of endogeneity 359
19.2 The problem with stochastic regressors 360
19.3 Reasons for correlation between regressors and the error term 363
19.4 The method of instrumental variables 367
19.5 Monte Carlo simulation of IV 369
19.6 Some illustrative examples 370
19.7 A numerical example: earnings and educational attainment of
youth in the USA 373
19.8 Hypothesis testing under IV estimation 378
19.9 Test of endogeneity of a regressor 379
19.10 How to find whether an instrument is weak or strong 381
19.11 The case of multiple instruments 381
19.12 Regression involving more than one endogenous regressor 384
19.13 Summary and conclusions 385
Exercises 387
20 Beyond OLS: quantile regression 390
20.1 Quantiles 391
20.2 The quantile regression model (QRM) 392
20.3 The quantile wage regression model 392
20.4 Median wage regression 396
20.5 Wage regressions for 25%, 50% and 75% quantiles 397
20.6 Test of coefficient equality of different quantiles 400
20.7 Summary of OLS and 25th, 50th (median) and 75th quantile
regressions 401
20.8 Quantile regressions in Eviews 8 402
20.9 Summary and conclusions 403
XII CONTENTS

Exercises 404
Appendix: The mechanics of quantile regression 405
21 Multivariate regression models 407
21.1 Some examples of MRMs 407
21.2 Advantages of joint estimation 408
21.3 An illustrative example of MRM estimation with the same
explanatory variables 409
21.4 Estimation of MRM 410
21.5 Other advantages of MRM 413
21.6 Some technical aspects of MRM 414
21.7 Seemingly Unrelated Regression Equations (SURE) 417
21.8 Summary and conclusions 419
Exercises 422
Appendix 424

Appendices

1 Data sets used in the text 425

2 Statistical appendix 436
A.1 Summation notation 436
A.2 Experiments 437
A.3 Empirical definition of probability 438
A.4 Probabilities: properties, rules, and definitions 439
A.5 Probability distributions of random variables 439
A.6 Expected value and variance 442
A.7 Covariance and correlation coefficient 444
A.8 Normal distribution 445
A.9 Student’s t distribution 446
A.10 Chi-square (2) distribution 447
A.11 F distribution 447
A.12 Statistical inference 448
Exercises 451
Exponential and logarithmic functions 455

Index 460
 PREFACE

The primary objective of this second edition of Econometrics by Example, as in the

first edition, is to introduce the fundamentals of econometrics without complicated
mathematics and statistics. The emphasis throughout the book is on explaining
basic econometric theory with several worked examples using data from a variety
of fields. The intended audience is undergraduate students in economics, business,
marketing, finance, operations research and related disciplines. It is also intended
for students in MBA programs and for researchers in business, government and
research organizations.

Major features
◆ In-depth examples illustrate major concepts in econometrics.
◆ Wherever essential, I have included figures and computer outputs from software
packages, such as Eviews (version 8), Stata (version 12) and Minitab (version 16).
◆ The data used in illustrative examples and in the exercises are posted on the book’s
website.
◆ Some of the exercises included are for classroom assignment.
◆ A full list of the data sets and the descriptions of the variables used in analysis is
provided in Appendix 1.
◆ Appendix 2 provides the basics of statistics that are necessary to follow this book.

New for the second edition

◆ There are two brand new chapters on quantile regression modeling and multi-
variate regression models. Two further chapters on hierarchical linear regression
models and bootstrapping are available on the book’s website.
◆ There are new illustrated examples in several chapters.
◆ I have considerably expanded the data-based exercises. In all there are about 70
data-based examples and exercises.
The book is now divided into five parts.
Part I discusses the classical linear regression model, the workhorse of economet-
rics, in considerable detail. These chapters form the foundation for the rest the
book. Some of the new topics discussed relate to the regression through the origin,
XIV PREFACE

or zero-intercept model, which is illustrated by the well-known capital asset pricing

model (CAPM) of financial theory, using UK stock market data. Another topic
included is the piecewise linear regression, in which linear segments of a regression
line are joined at certain break points, known as knots.
Part II examines critically the assumptions of the classical linear regression
model. Specifically, we discuss the topics of multicollinearity, heteroscedasticity,
autocorrelation, and model specification errors. We also discuss the topics of simul-
taneous equation bias and dynamic regression models. All these topics are discussed
with concrete economic data, some of which are new to this edition. The chapter
on heteroscedasticity discusses some technical aspects of robust standard errors and
introduces the concepts of the heteroscedastic robust t statistic and heteroscedastic
robust Wald statistic. The chapter on autocorrelation discusses both the standard
and alternative Durbin tests of autocorrelation. It also discusses the limitations of the
popularly used Jarque–Bera test of normality in small samples. Several data-based
exercises are added in this part.
Part III deals with what are known as the Generalized Linear Models (GLM). As
the name indicates, they are generalizations of the classical linear regression model.
Recall that the classical model assumes that the dependent variable is a linear func-
tion of the regression parameters, that it is continuous, that it is normally distributed,
and that it has a constant variance. The assumption of normal distribution for the
error term is for obtaining the probability distributions of the regression coefficients
and for the purpose of hypothesis testing. This assumption is very crucial in small
samples.
GLMs are useful in situations where the mean of the dependent variable is a non-
linear function of the regression parameters, the dependent variable is not normally
distributed, and the error variance may be non-constant. The GLMs discussed in this
part are: logit and probit models, multinomial regression models, ordinal regression
models, limited dependent variable regression models, and models of count data
using Poisson and negative binomial regression models. All these models are illus-
trated with several concrete examples
Some chapter specific changes include a discussion of the odds ratios in logit and
probit models and the bivariate probit model that involves two yes/no type depen-
dent variables that may be correlated. Chapter 11, on limited dependent variable
regression models, includes a discussion of Heckman’s sample selection model,
popularly known as the Heckit model. This part of the text includes several new
exercises, including a project for classroom assignment.
Part IV discusses several topics frequently encountered in time series data. The
concepts of stationary and nonstationary time series, cointegrated time series, and
asset price volatility are illustrated with several sets of economic and financial data.
The topic of economic forecasting is of great interest to business and economic fore-
casters. Various methods of forecasting are discussed and illustrated. As in the other
parts of this text, new examples and exercises are interspersed throughout this part
of the book.
Part V, which includes two chapters that are new to this edition, deals with some
advanced topics in econometrics.
Chapter 17 on panel data regression models shows how one can study the behav-
ior of cross-sectional units (e.g. firms in a given industry) over a period of time and
some of the estimation problems in such an analysis. One example is the impact of
 PREFACE XV

income and beer tax on the sales of beer in 50 US states and Washington, DC, over
the period 1985–2000.
Chapter 18 on survival analysis considers the time until an event occurs, such
as the time until an unemployed worker finds employment, the time that a patient
diagnosed with leukemia survives until death, and the time between divorce and
remarriage. We discuss in this chapter how econometric techniques handle these I
situations.
Chapter 19 on stochastic regressors and the method of instrumental variables
addresses a thorny problem in regression analysis, which is the correlation between
the regression error term and one or more explanatory variables in the model. If such
a correlation exists, the OLS estimates of the regression parameters are not even
consistent; that is, they do not converge to their true values, no matter how large
the sample is. Instrumental, or proxy, variables are designed to solve this problem.
The instrumental variables (IV) must satisfy two criteria: First, they must be highly
correlated with the variables for which they are a proxy but are not correlated with
the error term. Second, the IVs themselves must not be possible explanatory vari-
ables in their own right in the model for which they are acting as instruments. These
requirements are often not easy to meet, but in some situations IVs can be found.
Chapter 20 on quantile regression (QR) is new to this edition. Unlike the OLS focus
on estimating the mean value of the dependent variable in relation to one or more
explanatory variables, QR looks at the entire (probability) distribution of a random
variable by dividing the distribution into various segments, such as deciles, quartiles,
and percentiles. In skewed distributions or in distribution with several outliers, it
may be better to estimate the median of the distribution rather than the mean, for
the latter may be affected by outlying or extreme observations. In this chapter, I show
how QR estimates various quantiles and some of the merits of examining the whole
distribution. As a concrete example, we revisit the wage data and related variables
discussed in Chapter 1.
Chapter 21 on multivariate regression models (MRM) is also new to this edition.
MRMs are useful in situations in which we have more than one dependent variable
but each dependent variable has the same explanatory variables. An example is the
scholastic aptitude test (SAT) ,which most high school students in the US take. It has
two components: verbal and quantitative skills. One can estimate an OLS regression
of the test score on each skill separately, but it may be advantageous to estimate them
jointly, for the variables that affect each test scores are the same. It is thus quite likely
that the scores on the two tests are correlated. Therefore, joint estimation of the two
scores that takes into account the possible correlation between them will produce
estimators that are more efficient than if they are estimated separately by OLS.
However, if the errors are not correlated, joint estimation has no advantage over OLS
estimation of each equation singly.
A broader class of MRM is the seemingly unrelated regressions equations (SURE).
A classic example is the investment functions of several different companies in the
same industry group. Since these companies face a common regulatory atmosphere,
the investment decisions made by the individual companies may be estimated more
efficiently if we estimate them jointly rather than estimating each equation singly,
because it is quite likely that the error terms in individual regressions are correlated.
Note that, unlike the SAT example, in which the same individual takes both the
verbal and quantitative part of the SAT examination, in SURE that is not the case. In
XVI PREFACE

addition, in SURE the explanatory variables may be different for different companies.
Interestingly, if each company has identical explanatory variables, each taking identi-
cal values across every company, SURE estimates will be identical to those obtained
by estimating an OLS regression for each company individually. Also, if the error
terms across equations are not correlated, joint estimations of the equations has no
advantage over individual OLS estimation of each equation.

Companion website
The book’s companion website can be found at www.palgrave.com/companion/
gujarati-econometrics-by-example-2e/ and includes sources for both students and
instructors.
For the student, there are chapter summaries and conclusions, and all data sets in
Excel and Stata formats. Students are encouraged to use these data in several end-of-
chapter exercises to practice applying what they have learned to different scenarios.
A password-protected lecturers’ zone includes a collection of PowerPoint pre-
sentations that correspond to each chapter, and a Solution Manual with solutions to
all the end-of-chapter exercises. Because of their specialized nature, I have put two
additional chapters on the book’s website for those lecturers who wish to use them in
teaching: Chapter 22, Elements of hierarchical linear regression models, also known
as multilevel linear regression analysis (MLR), and Chapter 23, Bootstrapping: learn-
ing from the sample.

Hierarchical linear regression models

The primary objective of MLR is to predict the values of some dependent variable
as a function of explanatory variables at more than one level. A frequently studied
example is a child’s score on a standardized reading examination. It is influenced by
the characteristics of the child (e.g. the amount of study time) as well as features of
the child’s classroom (e.g. class size). This is a two-level analysis. If we include the
type of school, parochial or non-parochial, in the analysis, it would be a three-level
analysis. The actual level at which the analysis is done depends on the type of the
problem studied, availability of the data and computing facilities. As you can imag-
ine, the analysis becomes quickly more complex if we study a problem at many levels.
The point of this example is that we must consider the context in which the anal-
ysis is done. That is why MLR models are also known as contextual models. The
standard classical linear regression model is not adequate to deal with such multilev-
el analyses. The chapter on MLR explains the reasons for this and shows how such
multilevel regression models are estimated and interpreted.

Bootstrapping
In the classical linear regression model with the added assumption that the regression
error term is normally distributed we were able to estimate the parameters of the mod-
el, estimate their standard errors and establish confidence interval for the true param-
eter values. But what happens if the normality assumption is not valid or we have a
sample whose true population is unknown to us? In the chapter on bootstrapping, we
show how we can obtain the estimators of the parameters of interest, their standard
errors and the confidence intervals based on the computed standard errors.
 ACKNOWLEDGEMENTS

In preparing the second edition of Econometrics by Example, I have received invalu-

able help from Inas Kelly, Associate Professor of Economics, Queens College of
the City University of New York, and Professor Michael Grossman, Distinguished
Professor of Economics at the Graduate Center of the City University of New York.
I am indebted to them. I am also grateful to the following reviewers for their very
helpful comments and suggestions.

Reviewers of the first edition:

◆ Professor Michael P. Clements, University of Warwick
◆ Professor Brendan McCabe, University of Liverpool
◆ Professor Timothy Park, University of Georgia
◆ Professor Douglas G. Steigerwald, University of California Santa Barbara
◆ Associate Professor Heino Bohn Nielsen, University of Copenhagen
◆ Assistant Professor Pedro André Cerqueira, University of Coimbra
◆ Doctor Peter Moffatt, University of East Anglia
◆ Doctor Jiajing (Jane) Sun, University of Liverpool

Reviewers of the second edition:

◆ Professor Genaro Sucarrat, Norwegian Business School
◆ Doctor Jouni Sohkanen, University of St. Andrews
◆ Doctor Jin Suk Park, Durham University
◆ Professor Linus Yamane, Pitzer College
◆ Professor Doctor Horst Rottman, University of Applied Sciences Amberg Weiden
◆ Associate Professor Paul Solano, University of Delaware
◆ Professor Anh Nguyen, Ichec Brussels Management School
◆ Professor Frank J. Fabozzi, EDHEC Business School
◆ Professor Robert Duval, University of West Virginia
◆ Professor Robert Bickel, Marshall University
◆ Professor Nicholas Stratis, Florida State University
◆ Professor Giovanni Urga, Cass Business School and Bergamo University
XVIII ACKNOWLEDGEMENTS

and to the other anonymous reviewers whose comments were invaluable. Of course,
I alone am responsible for any errors that remain.
I am grateful to Jaime Marshall, Managing Director at Palgrave Macmillan Higher
Education for initiating this book and to Lauren Zimmerman, Development Editor
at Palgrave Macmillan, for her very constructive suggestions and for her meticulous
attention to detail in the preparation of the second edition. In addition, I am thankful
to Aléta Bezuidenhout and Amy Grant for their behind-the-scenes help.

The author and publishers are grateful to the following for kindly granting their
permission:
◆ Doctor Laurits R. Christensen and Doctor Douglas W. Caves for Table 21.6
Estimate of SURE airlines cost functions.
◆ Doctor Singfat Chu for Table 3.20 Diamond pricing.
◆ Professor Philip Cook for Table 17.11 The effect of beer tax on beer sales in 50 US
states and Washington DC, 1985–2000.
◆ Elsevier for Table 19.15 David Card’s OLS and IV wage regressions.
◆ Professor Ray Fair for Table 11.7 Data on extramarital affairs.
◆ Professor Philip Hans Franses, Professor Christiaan Heij, and Oxford University
Press for Table 8.13 Direct marketing of investment product.
◆ Edward W. Frees for Table 17.1 Charitable giving.
◆ Professor Jeremy Freese, Professor J. Scott Long and Stata Press for Table 12.7
Productivity of scholars.
◆ Professor James W. Hardin, Professor Joseph M. Hilbe and Stata Press for Table
8.12 Heart attack within 48 hours of myocardial infarction onset.
◆ John Wiley & Sons for: Table 7.22 Family planning, social setting, and decline in
birth rate in 20 Latin American countries, 1965–1975; the data used in Exercise
8.8; Table 8.9 The number of coupons redeemed and the price discount; Table
12.1 Data on R&D expenditure for 181 firms.
◆ Professor Leo Kahane for Table 5.1 Data on abortion rates in 50 US states for 1992.
◆ McGraw-Hill for the data used in Exercise 12.5.
◆ Professor Michael J. Kahn for Table 1.5 Data on 654 Boston youth.
◆ MIT Press for: Table 8.1 Data on smoking and other variables; Table 18.1
Modeling recidivism.
◆ Professor Tom Mroz for Table 4.4/Table 11.1 Married women’s hours of work and
related data.
◆ Professor Alicia Munnell for Table 17.9 Role of public investment in productivity
growth in 48 US states.
◆ NORC at the University of Chicago for the General Society Survey data used in
Exercise 9.1.
◆ Norton Company for Table 7.8 Consumption of cigarettes and death from lung
cancer.
◆ Professor Alan Reifman for Table 8.14 President Clinton’s impeachment trial.
 ACKNOWLEDGEMENTS XIX

◆ Professor Germán Rodríguez for the data used in Exercise 9.3.

◆ Sage Publications for the results reproduced in Exercise 8.8.
◆ Professor Allen L. Shoemaker for Table 3.21 Body temperature and heart rate.
◆ Professor C. F. Sirman, J. Shilling, U. Dhillon and John Wiley & Sons for Table 8.10
Fixed vs. adjustable rate mortgages.
I
◆ Standard and Poor for Table 10.8 Data on credit ratings of 92 US companies.
◆ Stata Press for Figure 19.1.
◆ Thomson Reuters DataStream for Table 2.15 CAPM of the UK stock market.
◆ Transparency.org and the World Bank for Table 2.18 GDP and corruption index.
◆ The UCLA IDRE Statistical Consulting Group for: Table 8.11 Admission to gradu-
ate school; Table 9.9 High school students’ curriculum choice; Table 12.8 Poisson
model of student absenteeism.
◆ The World Bank for Table 2.19 Fertility and related data for 64 countries.
 A PERSONAL MESSAGE FROM
THE AUTHOR

Dear student,
First, I am thankful to the students and teachers who used the first edition of
Econometrics by Example (EBE). Encouraged by their helpful comments and
suggestions, I have now written the second edition. It retains the user-friendly and
example-oriented approach to econometrics of the first edition. The changes I have
incorporated in this edition relate to some expository refinements of the topics cov-
ered in the first edition. I have added two new chapters in this edition and have put
two additional chapters on the book’s website. I have added several data-based new
exercises to the new edition.
As noted in the first edition, econometrics is no longer confined to economics
departments. Econometric techniques are used in a variety of fields, such as finance,
law, political science, international relations, sociology, psychology, medicine, and
agricultural science. Some techniques specifically developed for solving economic
problems have now found use in several of these disciplines. Newer econometric
techniques to address specific economic situations and refinements of old econo-
metric techniques are what keep the econometrics field an active field of study.
Students who acquire a thorough grounding in econometrics have a head start in
making careers in these areas. Major corporations, banks, brokerage houses, gov-
ernments at all levels, and international organizations like the IMF and the World
Bank employ a vast number of people who can use econometrics to estimate demand
functions and cost functions, and to conduct economic forecasting of key national
and international economic variables. There is also a great demand for econometri-
cians by colleges and universities all over the world.
There are now several textbooks that discuss econometrics from very elementary
to very advanced levels to help you along the way. I have contributed to this growing
industry with two introductory and intermediate level texts and this third book based
on a clear need for a new approach. Having taught econometrics for several years at
both undergraduate and graduate levels in Australia, India, Singapore, the USA, and
the UK, I came to realize that there was clearly a need for a book that explains this
often-complex discipline in straightforward, practical terms by considering several
interesting examples, such as charitable giving, fashion sales, pricing of diamond
stones, and exchange rates, in depth. This need has now been met with Econometrics
by Example.
What has made econometrics even more exciting to study these days is the
availability of user-friendly software packages. Although there are several software
packages, in this book I primarily use Eviews and Stata, as they are widely available
 A PERSONAL MESSAGE FROM THE AUTHOR XXI

and easy to get started with. Student versions of these packages are available at rea-
sonable cost and I have presented outputs from them throughout the book so you
can see the results of the analysis very clearly.
I have also made this text easy to navigate by dividing it into five parts, which are
described in detail in the Preface. Each chapter follows a similar structure, ending
with a summary and conclusions section to draw together the main points in an I
easy-to-remember format. I have put the data sets used in the examples in the book
up on the companion website, which you can find at www.palgrave.com/compan-
ion/gujarati-econometrics-by-example-2e/. Several data archives maintained by
academic journals and universities provide researchers a vast amount of data for
further research.
I hope you enjoy my hands-on approach to learning and that this textbook will be
a valuable companion to your further education in economics and related disciplines
and your future career.
I would welcome any feedback on the text; please contact me via my email address
on the companion website. Such feedback is extremely valuable to me in planning
future editions of this book.
 LIST OF TABLES

Tables marked with an * are available on the companion website: www.palgrave.com/companion/

gujarati-econometrics-by-example-2e/
Table 1.1 Wages and related data*
Table 1.2 Wage regression 15
Table 1.3 Stata output of the wage function 17
Table 1.4 The AOV table 17
Table 1.5 Data on 654 Boston youth*

Table 2.1 Production data for the USA, 2005*

Table 2.2 Cobb–Douglas function for USA, 2005 30
Table 2.3 Linear production function 31
Table 2.4 Cobb–Douglas production function with linear restriction 33
Table 2.5 Data on real GDP USA, 1960–2007*
Table 2.6 Rate of growth of real GDP, USA 1960–2007 35
Table 2.7 Trend in real US GDP, 1960–2007 36
Table 2.8 Food expenditure and total expenditure for 869 US households in 1995*
Table 2.9 Lin-log model of expenditure on food 38
Table 2.10 Reciprocal model of food expenditure 39
Table 2.11 Polynomial model of US GDP, 1960–2007 41
Table 2.12 Polynomial model of log US GDP, 1960–2007 42
Table 2.13 Summary of functional forms 43
Table 2.14 Linear production function using standardized variables 45
Table 2.15 CAPM of the UK stock market, monthly data for 1980–1999*
Table 2.16 Market model of the UK stock market 48
Table 2.17 The market model with intercept 48
Table 2.18 GDP and corruption index*
Table 2.19 Fertility and related data for 64 countries*

Table 3.1 A model of wage determination 54

Table 3.2 Wage function with interactive dummies 56
Table 3.3 Wage function with differential intercept and slope dummies 57
Table 3.4 Reduced wage function 58
Table 3.5 Semi-log model of wages 61
Table 3.6 Gross private investment and gross private savings, USA, 1959–2007*
Table 3.7 Regression of GPI on GPS, 1959–2007 62
Table 3.8 Regression of GPI on GPS with 1981 recession dummy 63
 LIST OF TABLES XXIII

Table 3.9 Regression of GPI on GPS with interactive dummy 63

Table 3.10 Quarterly retail fashion sales 1986-I–1992-IV*
Table 3.11 Results of regression (3.10) 65
Table 3.12 Sales, forecast sales, residuals, and seasonally adjusted sales 67
Table 3.13 Expanded model of fashion sales 68
Table 3.14 Actual sales, forecast sales, residuals, and seasonally adjusted sales 69 I
Table 3.15 Fashion sales regression with differential intercept and slope dummies 70
Table 3.16 Hypothetical data on production lot size and average cost (AC,$) 72
Table 3.17 Relationship between average cost and lot size 72
Table 3.18 Lot size and average cost relationship disregarding the threshold value 73
Table 3.19 Effects of ban and sugar consumption on diabetes*
Table 3.20 Diamond pricing*
Table 3.21 Body temperature, gender, and heart rate*
Table 3.22 A sample of 528 wage earners*

Table 4.1 The effect of increasing r23 on the variance of OLS estimator b2 82
Table 4.2 Hypothetical data on expenditure, income, and wealth for 10 consumers 83
Table 4.3 Regression results (t values in parentheses) 83
Table 4.4 Mroz data on married women’s hours of work*
Table 4.5 Women’s hours worked regression 85
Table 4.6 The VIF and TOL factors 87
Table 4.7 Revised women’s hours worked regression 88
Table 4.8 VIF and TOL for coefficients in Table 4.7 89
Table 4.9 Principal components of the hours-worked example 90
Table 4.10 Principal components regression 91
Table 4.11 Manpower needs of the US Navy*
Table 4.12 Data on blood pressure and related variables for 20 patients*
Table 4.13 Longley classic data*

Table 5.1 Data on abortion rates in the 50 US states for 1992*

Table 5.2 OLS estimation of the abortion rate function 98
Table 5.3 The Breusch–Pagan test of heteroscedasticity 101
Table 5.4 Abridged White test 103
Table 5.5 Transformed Eq. (5.1) 104
Table 5.6 Logarithmic regression of the abortion rate 105
Table 5.7 Robust standard errors of the abortion rate regression 106
Table 5.8 Heteroscedasticity-corrected robust standard errors of the wage function 109
Table 5.9 Heteroscedasticity-corrected robust standard errors of the hours function 109
Table 5.10 GDP growth rate and related data for 106 countries*
Table 5.11 Data for 455 US manufacturing industries, 1994*

Table 6.1 US consumption function, 1947–2000*

Table 6.2 Regression results of the consumption function 114
Table 6.3 BG test of autocorrelation of the consumption function 120
Table 6.4 First difference transform of the consumption function 122
Table 6.5 Transformed consumption function using = 0.3246 124
Table 6.6 HAC standard errors of the consumption function 125
Table 6.7 Autoregressive consumption function 126
XXIV LIST OF TABLES

Table 6.8 BG test of autocorrelation for autoregressive consumption function 127

Table 6.9 HAC standard errors of the autoregressive consumption function 128
Table 6.10 Housing starts and related data for USA, 1973–2011*

Table 7.1 Determinants of hourly wage rate 133

Table 7.2 Expanded wage function 133
Table 7.3 Refinement of the wage model 134
Table 7.4 RESET test of the wage model 136
Table 7.5 The LM test of the wage model 137
Table 7.6 Regression of experience on age 139
Table 7.7 Determinants of log of wages 140
Table 7.8 Consumption of cigarettes and death rate from lung cancer in 11 countries 143
Table 7.9 Deaths from lung cancer and consumption of cigarettes (all countries) 143
Table 7.10 Deaths from lung cancer and consumption of cigarettes (USA excluded) 144
Table 7.11 Calculated p-value equivalents to true alpha levels at given sample sizes 146
Table 7.12 Aggregate consumption function for the USA, 1960–2009*
Table 7.13 Reduced form regression of PCE on GDPI 151
Table 7.14 Reduced form regression of income on GDPI 151
Table 7.15 OLS results of the regression of PCE on income 152
Table 7.16 OLS results of regression (7.22) 157
Table 7.17 Results of regression with robust standard errors 157
Table 7.18 The results of regression (7.23) using HAC standard errors 158
Table 7.19 OLS estimates of model (7.26) 161
Table 7.20 OLS estimates of model (7.26) with HAC standard errors 161
Table 7.21 Cigarette smoking and deaths from various types of cancer in 43 US states and
Washington, DC, 1960*
Table 7.22 Family planning, social setting, and decline in birth rate in 20 Latin American
countries, 1965–1975 165

Table 8.1 Data on smoking and other variables*

Table 8.2 LPM model of to smoke or not to smoke 171
Table 8.3 Logit model of to smoke or not to smoke 176
Table 8.4 The effect of a unit change in the mean value of the explanatory variables on the
probability of smoking 178
Table 8.5 The logit model of smoking with interaction 179
Table 8.6 Odds ratios for smoking versus non-smoking 181
Table 8.7 Probit model of smoking 182
Table 8.8 The probit model of smoking with interaction 183
Table 8.9 The number of coupons redeemed and the price discount 185
Table 8.10 Fixed vs. adjustable rate mortgages*
Table 8.11 Admission to graduate school*
Table 8.12 Heart attack within 48 hours of myocardial infarction onset*
Table 8.13 Direct marketing of an investment product*
Table 8.14 President Clinton’s impeachment trial*

Table 9.1 Data on school choice*

Table 9.2 Multinomial logistic model of school choice 195
Table 9.3 Raw data for mode of travel*
 Another Random Document on
Scribd Without Any Related Topics
Her tone was that of a woman who intended to stand no further
nonsense of any kind. She was dog-tired, already angry on Joan's
account, and the resigned and hopeless air of the slender creature
before her completed her resolve to get to the bottom of things.
But nothing appeared to be happening about Audrey and Monty.
Mrs. Cunningham was still of the same mind, or no mind, that had
prompted her letter.
"Where is he now?" Mollie demanded.
Audrey did not know.
"Is he with Philip?"
She did not know that either.
"And aren't you going to Lennox Street?"
"I don't think I can."
Mollie's eyes went round the room. I myself have never been in Mrs.
Cunningham's bed-sitter, but I have been in many others and can
picture its frugality—the gas-ring, the little cupboard with the bread
and jar of marmalade in it, the chintz-covered grocer's box on which
Mollie's cup of cocoa stood, probably a slightly fresher wallpaper-
pattern where the Jacobean wardrobe had stood. Lennox Street was
positive luxury by comparison, yet here was Audrey preferring this.
Then there was Audrey herself, heavy-eyed, drained of energy,
probably thinking of George Cunningham and wondering whether
any experience was worth repeating. With Joan for breakfast and
Audrey for supper, poor Mollie had had about enough of it for one
day.
"About this marriage," she said abruptly. "Of course, it's not going to
be put off. I shall see to that. I shall have a talk with Monty too
when I've finished with Philip. You're simply run down and want a
tonic."
"Oh, it isn't that," Audrey replied, sinking into an old wicker chair.
"I've thought it all over. I don't think they tell young girls enough
before they get married. They ought to tell them lots more. They
ought to tell them quite plainly, 'You'll have to be prepared for this
and that and the other. You'll have to expect to sit up half the night
in the dining-room with dinner on the table wondering where he is.
You'll have to learn that he hasn't really been run over or anything of
that kind and that it's only their way. You must expect telegrams and
telephone calls and excuses, and you mustn't be surprised if he
brings somebody else with him when he does come. They're like
that. And when they come home in a beastly state——'"
Here Mollie peremptorily interrupted her.
"Leave that brute in his grave," she commanded. "We're talking
about Monty, not him. And I'll see Monty. Now what's all this
rigmarole about milkmen and cellars and all the rest of it? Tell me as
I undress you. I'm going to put you to bed."
But little that was fresh was to be learned here either. Audrey
thanked her again and again for the offer of the house, but she
thought she would rather be here with her gas-ring and cocoa and
chintz-covered sugar-box. Mrs. Cook thought she could arrange it—it
would only be half a crown more.
"Well, I'll see Mrs. Cook too while I'm about it; may as well do the
thing thoroughly. Let me unlace your boots—why, your feet are cold,
and on a night like this! Never mind your hair; you can do it in bed.
And drink the rest of this cocoa. Really I do think I live in a helpless
sort of world—there isn't enough of me to go round—there ought to
be half a dozen of me. Now into bed with you, and you'd better stay
there till I come round in the morning. I'm not going back till
Wednesday."
She packed up the cocoa-cups and turned off the gas-ring, opened
the window and wound up the little Swiss clock. As she moved about
the room folding Audrey's clothes and setting things to rights her
own letter of Sunday morning was brought up, but she placed it on
the mantelpiece by the side of the clock, forbidding Audrey to read it
till the morning. Letters didn't matter now that she was here in her
own capable, practical person. Letters took too long. She was going
to have things done much more quickly or know the reason why.
 XIII
Philip himself opened the door to her. She gave her cheek to be
kissed and then walked straight in.
"Is Monty here?" were her first words.
"No. I think he's gone out for a walk."
"A walk, at this time of night!"
Philip shrugged his shoulders. "What brings you here, Mollie?" he
said.
She was busy untying her veil. "What do you suppose? Everything,
of course."
"Have you had dinner or supper?"
"I had a cup of cocoa at Audrey Cunningham's, and don't want
anything else. Now why didn't you answer my telegram, and why
didn't you write again as you promised?" She threw the hat and veil
on the table and her gloves after them and stood before Philip.
"How are the children?" he asked.
"Perfectly well."
"And Joan?"
Her only answer to this last was a long look. Then she walked to the
little Empire sofa and sat down. He might stand if he wished.
"Well?" she said at last. This was after a full minute, during which
time he had stood by the table idly fingering her veil. Then suddenly
his fingers pushed the veil aside, and he crossed to the sofa and sat
down by her.
"What is it you want to know, darling?" he asked.
"Everything—every little thing from beginning to end," she replied.
"You've seen that I don't want to tell you?"
"Yes, I've seen that."
"And that I probably have my reasons?"
"Oh—reasons!"
"Reasons that are stronger than ever at this moment?"
"Will they go on getting much stronger? If so I can only warn you
that a breaking-point will come."
"Yes. I'm very near it."
"And so am I, Philip."
There was no mistaking her tone. It did not mean that if he
continued to shut her out like this she would do anything violent—
live apart from him, become merely his housekeeper, or anything of
that kind. It meant enormously more than that. Where confidence
and trust are, there are few divergences that do not presently right
themselves, few differences that cannot be resolved; but where
these are absent nothing is right. Every word is possible peril, every
silence a hanging sword. In all my acquaintance I know of no
happier marriage than the Esdailes'. You never go into their house
and feel that the air is still charged with some scene that your arrival
has interrupted, you never leave wondering what weapons will be
picked up again the moment your back is turned. Philip is not
without his tempers nor Mollie without her own purposes, but it
stops at that. The rest is brave decencies, with I know not what
tenderer stuff behind. This it was that seemed for the moment to be
in peril.
But suddenly she put her hand on his. She did not speak; the hand
spoke for her. The next moment his other hand had fallen on hers
again, so that both enclosed it. Then their eyes too met.
"It would be an awful thing to risk, Phil," she said quietly.
His eyes begged her. "Won't you let me carry it a little longer alone?"
"I don't believe you can. And if you can," she added, "I don't see
what we got married for."
"But it will be rotten for you too."
"When have I shrunk from that?" she asked.
"Never," he replied in a low voice.
And so Mollie Esdaile too took her portion of the burden.
"Well, where shall we begin?" he asked, with a sigh that it must be
so.
"You know best. But tell me first why you didn't write."
"I hadn't seen young Smith. I haven't yet as a matter of fact. But
——," he drew her head to his breast, and there were some
moments during which he whispered into her ear.
For all his care and guarding it was not possible that she should not
at least tremble. But she did not start within his arms. The tremor
passed, and her dry lips repeated—
"Shot him!"
"For some reason or other. I don't think either of us can quite tell
Joan that, can we?"
"Shot him! Chummy!"
"There's no doubt about it. It was Monty who picked up the pistol.
Neither you nor I can very well tell Audrey that, can we?"
"Monty found the pistol!"
"And I picked up the cartridge-shell myself. The police were round
here at six o'clock on Friday morning looking for them."
"The police! Here!"
In spite of all, Philip could not restrain a little laugh. "Oh, they didn't
find anything. They don't when they've given you warning the night
before. By that time both pistol and case had been at the bottom of
silver-flowing Tamesis for six hours. I dropped 'em in myself from
the middle of Albert Bridge. Do you begin to see what you're in for,
my poor darling?"
It was doubtful whether she yet did. She could still only repeat,
"Chummy shot him! Does that mean——?" Her horrified stare
finished the question.
"Oh, I don't think so," he answered quickly, "at any rate not yet.
Naturally both Hubbard and I have got to stand by Chummy for the
present."
"Then somebody saw Monty pick up the pistol?"
"For the police to know anything about it, you mean? Well, as a
matter of fact that's the purest bad luck. There happens to be a
fellow called Westbury, confound him, and that beastly bullet seems
to have fetched up somewhere in his house; he lives just across the
way there. That's all the police have to go on now that the other
things are safely in the river mud."
Slowly it was sinking into her mind. Her eyes closed for a moment as
she felt the first faint strain of the weight of it. This came with the
thought of Joan.
"But—but——" she said faintly, "—that poor child——?"
"Joan? I suppose she is wondering what's happened?"
"Wondering what's happened.... You may as well know, Philip—it's a
mere trifle at this stage—that they're most awfully thick—far more
than you've been told. You know what boys and girls are nowadays.
And now comes this horrible silence, except for that one little note
from him——"
"Has she had a note from him?" Philip asked quickly. "From the
hospital?"
"There was no address on it. I suppose it was from the hospital?"
she in her turn asked in quick alarm.
"Yes, it wasn't from a prison. When did she get this note?"
"Yesterday morning."
"Did she show it to you?"
"Yes. It simply said he'd had this spill, but was all right, and she
wasn't to be alarmed."
"He didn't mention incidentally that he'd shot a man?"
"Of course not. I don't believe he had."
Philip passed this last point.
"Well, as he's written once I suppose he'll go on writing. He's heaps
better as a matter of fact. So it won't be too hard on her. Anyway,
she'll have to grin and bear it."
"I did hold out hopes that I might be able to take him back with
me," Mollie ventured.
"You won't be able to do that," said Philip with decision. "Quite apart
from his being fit to travel, we've only gone a certain way in all this,
you know. He'll still have matters to explain. And till he does, I'm
afraid Audrey Cunningham will have to make the best of things too,
like Joan. It wouldn't do her any good to know that Monty was liable
to arrest at any moment."
"But is he?" said Mollie, startled.
"Of course he is. So am I. So are you. So would she be."
"But why?"
In spite of his explanation, I don't think she understood. I don't
think she understands to this day. I don't think that at the bottom of
her anti-social heart any woman does. A delayed wedding or a post
without a lover's letter is a far greater thing than a capital charge in
which all who conspire are principals.
Then, in spite of her fatigue, her skeptical common sense came to
her aid. Philip might involve himself in a web of unelucidated stuff of
which one-tenth perhaps was fact of sorts and the rest pure
speculation; but she knew Chummy. The thought of Chummy as a
murderer was absurd beyond words. Whatever the explanation
might be it certainly was not that. And, yawning as she rose, she
told Philip so.
"And that's that," she concluded. "Now do let's go to bed. Of course,
if you think Chummy's a murderer I quite see why you didn't write
and why you don't want to tell Audrey and all the rest of it, but you'll
find it's all a mistake. There's something you don't know, or else
there's been an accident of some kind. If you seriously want me to
believe that Chummy Smith.... What's the matter, darling?"
The last words were a quick, startled cry. She did not know what it
was that lurked at the bottom of the eyes that were looking so
deeply and somberly into her own, but she feared already. His head
was slowly shaking from side to side.
"Philip! What do you mean?" she cried in agitation.
Still the head shook. It was impossible for her mind not to fly back
to that moment, now nearly five days ago, when he had stood
blinking in the doorway with a candle in one hand and a jar of
liqueur in the other.
"Tell me quickly what you mean, Philip!" she cried again.
"I saw it."
She fell back. "You——?"
"It wasn't an accident, and there isn't anything I don't know."
"You——?"
The slow sideways shake changed to one single nod. The next
moment his arms were about her and he was leading her to the sofa
again. He sighed. There was no help for it now. If she would have it
she must have it all.
"It's the only thing I haven't told you. We may as well get it over,"
he said.
Nor did he whisper this time. He spoke in his usual voice, using the
plainest English he could.
But what it was that Philip Esdaile told his wife you must guess for a
little while longer. She was the first living soul to know. And it was a
very different thing from that which she had left Santon to hear.
For it was this overwhelmingly extraordinary yet stupendously
ordinary thing that sent her round to Audrey Cunningham the next
morning, but without comfort for her, with no plans for settling the
wedding out of hand.
It was this same thing that took her back to Santon on the
Wednesday, without Chummy, without help for Joan.
It was this same thing that puzzled Monty Rooke's brain as he took
his midnight walk that night down Roehampton Lane, driven from
Audrey Cunningham's company and sick of the sight of Philip and all
his works.
And it was this and nothing else that Cecil Hubbard so much wanted
to know when he knitted his honest brows over hydrophones,
sound-ranging, or whatever other mysterious apparatus it was that
Philip Esdaile might have hidden away in his cellar.
 PART IV

THE MAN IN THE PUBLIC-HOUSE

 I
As an eager and passionate student of the Life of my day there are,
within limits, few places that I don't visit and few people I don't on
occasion talk to. I say "within limits," since I admit that there may
be grades at one end of the scale at which I draw the line, while at
the other end there may conceivably be those who draw the line at
me. But within these extremes, if not always familiarly, yet on the
whole without constraint, I sup at coffee-stalls or dine in quite good
company more or less indifferently.
I have found that the best strategic jumping-off-points for the
satisfying of this curiosity about the preponderating average of Life
are two. One—the Public-house—I have already mentioned. There
only a glass screen may divide you from the hawker who has left his
barrow for a few minutes in somebody else's charge, or from the
gibused and silk-mufflered figure who finds a glass of sherry a
convenient way of getting small change for his taxi. The other point
of vantage is the Club, where that same taxi is paid off, but where
liveried chauffeurs may stand for hours by the waiting cars.
I shall come to the Man in the Club by and by. For the present I wish
to return to the Man in the Public-house.
I won't say that I always love him, but I always recognize that I
have him very seriously to deal with. I am not thinking of him now
either as a reader of my journalism or as a potential buyer of my
novels, but as a larger phenomenon. I am thinking of him—loosely I
admit—very much as some political cartoonist might think of a
generalized and consolidated figure that turns a deaf ear to the
Bolshevist and his sinister whisperings on the one side, while the
other ear is no less stopped to the honeyed blandishments of the
statesman who so frequently and extraordinarily seeks to cajole him
with flatteries that are both out-of-touch and out-of-date.
He is Conservative, if Conservatism means that he cynically holds his
hand till he has seen what the next dodge is likely to be; and he is
Liberal in the sense of believing that if everybody looks after himself
then there will not be anybody who is not looked after. You have
overdone it, my good friends and representatives in Parliament. He
no longer believes a word you say. You offer him good and
necessary things, and he glances sideways at you, and his lips shape
the words "By-election." You try to keep him from rash and
dangerous courses, and he wants to know what you are getting out
of it. You ration him, but he knows where to get sugar and butter
while you make the best of saccharine tablets and West African
margarine; you de-control, and he knows better than you do why
hens cease to lay and rabbits to breed. It is we of the cheaper Press
who really have him in hand, and he cocks his ears back at us
occasionally. He did so when the Daily Circus gave him pictures of
bathing girls instead of war news; he is doing so to-day when—oh,
lots of things—are pushed on and off the proscenium like Monty
Rooke's camouflage canvas trees and linoleum sentries. He sniffs at
all these things, says nothing, and calls for another glass of his
country's seven-times-accursed beer.
Yet he follows, if not his leader, his neighbor. This he sometimes
does to the most astonishing conclusions, just as he looks up at the
sky-line because he sees somebody else doing so, or is prepared to
swear that he hears a maroon because the man next to him says
"Listen!" The vaguer the rumor the greater is the scope for his self-
and-collective suggestion. He sees Russians, he knows that dead
Field-Marshals are still alive. He can tell you from private information
that such-and-such a battalion has been cut up or such-and-such a
battle-fleet sent to the bottom of the sea. And not one of these
larger things is half so large to him as the smaller thing that looms
huge because it is in his own immediate neighborhood. We on the
Circus provide pictures to give him at least the photographic
semblance of body for his belief. But what when the very flesh and
blood of the drama passes along his street every day? What when
he has spoken with the chief actor himself, knows who he married,
the number of his children and their names? What when he knows
the house he lives in, who lived there before, why he left so
suddenly, and the very words he is said to have said on leaving?
Do you see what I am coming to—those first faint whisperings of
something wrong—or if not positively wrong so much the better for
public-house debate on the point—with Philip Esdaile's house in
Lennox Street?
 II
The first that Esdaile knew of all this was from the younger of
Mollie's two maids. Monty and Audrey had arranged to dispense with
the services of these two domestics, but Philip, still lingering on, had
wanted the younger one at least back. She had promised to come,
but had not done so, and Philip had sought her out. Thereupon she
had said that she would rather not come.
"Why?" Philip had asked; but she had given no satisfactory reason.
He had then turned to the second maid, but with no better result.
After all, it didn't matter. It was very little trouble for himself and
Monty to make their own breakfasts. They could take their other
meals at the Chelsea Arts Club, and there would be no difficulty in
getting a woman one or two days a week to clean.
Then, to his extreme astonishment, on the very day after Mollie's
departure for Santon, he left Monty to a sandwich-and-coffee
luncheon in the studio and came out of his house to find Lennox
Street almost as full of people as it had been on the morning when
the parachute had descended on the studio roof.
"What's the matter? Anything happened?" he asked the nearest
loiterer at his gate; but he did not learn what had really happened
till he reached the Club.
Certainly the joke, if it was a joke, appeared to be "on him."
Simultaneously two grinning fellow-members thrust into his hand
that morning's issue of the Roundabout. The Roundabout, I should
say, is the Circus's (much inferior) rival.
It contained a photograph of Esdaile's house, with the spot where
the parachute had descended marked with a cross.
It was, of course, a thousand pities. No man likes the house he lives
in held up to the idle public gaze. Had the annoying thing been
submitted to my own paper I could have stopped it. Had it been a
big thing I might even have stopped its appearance in the
Roundabout, for, while we cut one another's throats in detail, we
have our understandings in larger matters. Hurriedly I scanned the
rest of the paper to see whether any letterpress went with the
picture. None did. There was simply the photograph, with a couple
of quite innocent descriptive lines underneath.
"Seems to me rather a stumer," I said to Willett. "Is Hodgson losing
his grip a bit?"
"Haven't noticed it," Willett replied. "Sound man Hodgson. Doesn't
often do things without a reason. I think we might go a bit slower on
actresses and mannequins. This is the crash we were talking about
the other morning, isn't it?"
"Yes. A wash-out I should have said."
"Perhaps he's playing the local-interest card. He's doing that just
now. I don't see why we shouldn't do more of it."
"I think we'll wait for a better story than that anyway," I replied.
"Well, let's get to work——"
But all that afternoon the thing worried me. It was a trifle, perhaps,
but it was a trifle on the wrong side. More, unlike some other trifles,
I already saw how dangerously capable of further development it
was. I have told you what the attitude of the Press was to this
question of civil flying. It was one of simply awaiting events. But all
the time events were fermenting, so to speak. High over our heads
Olympian minds were shaping and re-shaping policies and plans,
and Argus eyes were tirelessly watching for indications of the
receptivity of the popular mind. Had Hodgson heard something that
we had not? As you sometimes see an insignificant person's affairs,
of no interest in themselves, solemnly weighed by the Lords of
Appeal because of some novel and far-reaching point they raise, was
something in the nature of a Test Case now being sought? Had we
on the Circus been wrong in assuming that the idea was simply to
catch and make an example of the careless joy-rider and the idiot
who stunted over towns? Was some more important point to be
raised, and had Hodgson had wind of it?
I was inclined to think not, and for the reason I have just given.
Make a thing big enough, and we hang fairly well together; but take
the whips off, so to speak, and we go as we please. If it had been as
important as all that we should have heard of it. Willett, who is a
youngster of parts, was in all probability right. Hodgson was merely
catering for the local interest.
But still I was uneasy, and my uneasiness had nothing to do with the
annoyance the publication of the photograph of the house in Lennox
Street must cause Esdaile. I was thinking of far graver possible
consequences. Even the lightest measure of Publicity is not a thing
to be trifled with. Here I know what I am talking about. The merry
fellows of the Chelsea Arts Club might pull Esdaile's leg about his
haunted house, and want to know whether the White Lady dropped
any hairpins as she passed, or if the horrible shrouded figure with
the crimson-dripping hands would make a good film; but we
journalists have to take these things a good deal more seriously than
that. Publicity, sometimes of the most incredibly silly kind, is our
meat and drink and hourly breath. All day and every day our brains
are on the stretch in our endeavors to secure it. We bring our
heaviest guns to bear on the elusive thing, are sure we can't
possibly miss it this time, let fly, and lo! we have missed after all.
Like a pithball on a fountain, it is still dancing there untouched, and
any penny peashooter may bring it down when all our trained
intelligence has failed.
And what would be the effects on our Case if it came down?
Well, you can see that for yourself. In obscurity lay our hope that the
thing might remain what on the face of it it appeared to be. Switch
the arc-lamps of the great papers on to it, with the whole power-
house of dynamic government behind them, and all was over. Not an
aspect of the Case would go unprobed to the very bottom, and the
hungry newspapers would find themselves, not with a mere
aeroplane crash that could be dismissed in a couple of lines, but with
a really fine fat, first-class Murder Case that would keep them
merrily going for weeks.
And I can assure you that we all wanted very badly indeed just such
a Case. We wanted it for more reasons than one. We wanted it, as
we always do, in the ordinary way of our business, but much more
we wanted it to take people's minds off other matters. We wanted it
for the same reason that made us resolutely print those pictures of
girls bathing during the blackest days of the War. We wanted it
because the Man in the Public-house was restless and showed a
disposition to pry into affairs in which his interference is only wanted
when a General Election draws near. Bathing girls were very well in
their way; a really high-class line in Divorce Cases would have
outstripped them easily, if I may be permitted the unintentional
expression; but the man who could have given us on the Circus the
first Assassination in the Air could have named his own price for it.
 III
The flat in which I live with the old housekeeper who looks after me
is not in Chelsea at all, but a quarter of an hour's walk away, just
round the corner from Queen's Gate. It is exceedingly comfortable
(as indeed it should be considering the rent I am made to pay for it),
I have my own furniture, and on the whole I don't ask for a much
better place to work in. For, quite apart from my paper, I do work,
and I don't want to give you the impression that the whole of my
leisure time is given over to the investigation of what happens to my
Chelsea friends.
I was, as a matter of fact, particularly busy just about that time. Day
after day I was getting up at half-past six in the morning,
breakfasting at my table as I worked, and continuing without
interruption till it was time for luncheon and the office. Since you are
probably not in the white-elephant line of business, I won't tell you
which of my novels I was at work on. I will only say that I at any
rate was interested in it, and, severe as was the strain of writing
from seven o'clock in the morning till midday, I sometimes hated to
break off. Mrs. Jardine had orders not to admit anybody whomsoever
between those hours, and obeyed them to the letter.
You may judge then of my surprise when there walked into my study
at nine o'clock in the morning, and not over Mrs. Jardine's dead
body, Billy Mackwith.
"Don't scold the old lady," he began without preface. "I suppose she
hadn't any barbed wire except that on her chin—it is rather like one
of the gooseberries we used to make on the old wiring-course. I had
to see you."
"Had breakfast? Have a cup of coffee?" I asked him.
"Nothing, thanks. Well, I think I saved friend Philip a certain amount
of trouble yesterday," he said, putting down his hat, stick and
gloves. I don't think Mackwith buys a glossy new silk topper every
time he goes out, but I do honestly believe he buys a new pair of
lemon-colored gloves.
"Oh? How was that?"
"At the inquest on that fellow," he replied. "And by the way, I saw
the Roundabout too. I suppose it has its humorous side, but it's very
annoying too. I should go for 'em for libel. A house can be libeled,
you know. Anyway, it's a good job he's out of town."
I was on the point of saying, "A good job he's what?" when I
checked myself. If you remember, I had last seen William Mackwith,
K.C., when I had left him at Sloane Square Station an hour or two
after that confounded aeroplane accident. He and Hubbard had gone
off to keep their respective appointments, while I myself had
followed our check-coated friend Westbury into a public-house.
Whether either Esdaile or Hubbard had seen him since I didn't know.
I now gathered that Billy at any rate hadn't seen Esdaile.
"Yes?" I said. "What trouble have you been saving him?"
"Well, I told you—I saved him the bother of stopping for the
inquest."
I had of course known that there must be an inquest, but I suppose
I had been busy and had forgotten it again. This began to be
interesting.
"Tell me about the inquest," I said.
He took a cigarette from his case and offered me one. Then he
continued between puffs.
"Well, as a matter of fact there wasn't very much trouble. One man
seemed inclined to be cantankerous, but we brought it in
Misadventure all right. We——"
I imagine that at this point he caught sight of the expression on my
face, for he stopped suddenly.
"Esdaile did go out of town, didn't he?" he asked.
"I'll tell you about Esdaile in a moment. Go on about the inquest."
He seemed puzzled, but went on.
"Of course, that was my idea in speaking to that Inspector that
morning. It would have been a pity to upset the Esdailes' plans. So I
explained this to the Inspector and gave him my card—said I hadn't
seen very much, but as much as anybody else—result, I was made
foreman of the jury."
Here I had a little flash of illumination as regards Inspector Webster
too. Esdaile, if you remember, had said to him, "Yes, that was Mr.
Mackwith, the King's Counsel; didn't you know?" and Webster had
answered, "Was he indeed, sir?" My respect for the Inspector's
powers of giving nothing away went up several hundreds per cent.
Apparently it was the Inspector who had seen to it that Billy had
been put on the jury.
"Well, you were made foreman. You said one man gave trouble.
Who, and why?"
"Oh, some fellow or other—Westcott or Westmacott I think his name
was—I forget. Insisted on viewing the body. Wanted his money's
worth I suppose. He was sorry he did though."
This was more and more interesting. I asked what sort of a man this
Westmacott was.
"The sort of fellow who would be down in the cellar before his wife
and children when there was an air-raid on, I should say," Billy
replied. "Awful nuisance of a man. But he got his all right. He'll
probably be taking solid food again this day week."
"Then you did see the body?"
"Had to, if only to keep this fellow quiet. He stuck out right to the
finish too, but we got our dozen without him. Prima facie case, of
course. Death by burning, and what wasn't that was general smash-
up."
"Was a doctor called?"
"The divisional surgeon was there, but he quite agreed, and I saw to
the rest in my capacity as foreman. There was only one man who
wasn't satisfied, and he was busy——" Billy twinkled wickedly.
You may imagine how I was beginning to relish all this. The Chelsea
Arts with its rags about haunted houses and White Ladies who
dropped hairpins was well enough in its way, but its humor could not
compare for a moment with the spectacle of a rising King's Counsel
who practically forced himself on to a jury-panel, got himself made
foreman, and then burked inquiry by shutting up the only juryman
who had as much as a suspicion of the dangerous truth—and all this
in the whitest innocence and purest good faith! I could have laughed
aloud. Had he been a willing instrument in the affair he could not
have done his work more efficiently and completely.
"Is the poor fellow buried yet?" I asked in a suppressed voice.
"Yesterday," said Billy.
"And he can't be dug up again?"
Mackwith gave me a sharp look. "What do you mean?" he asked
quickly.
"Dug. Past participle of the verb to dig. I mean is he buried once for
all?"
"Short of an Exhumation Order from the Home Secretary he is. I
don't understand you."
"And that's rather difficult to get, isn't it?" I continued.
"I should say the North Pole was comparatively easy," Billy replied.
At this point my laughter really became too much for me. I
remember hoping that it didn't seem too rude, but I couldn't help it.
Billy let me finish, and then asked quietly, even gravely, "Now if
you're feeling better, will you please explain?"
I suppose my laughter had been just a little hysterical. As I have
already told you, I myself stand only on the verge of this Case; but
not so my friends, and Esdaile in particular. I remembered—and
deep under that rather remarkable laughter it moved me more than
a little to do so—the extraordinary range and sweep of emotions
that had shaken Esdaile in the course of a single day. I remembered
the bright strained tension of those first minutes after he had come
up out of the cellar, the shock of that telephone message from the
hospital that had told him what had happened to a friend. I
remembered that black depression when Hubbard and I had found
him waiting alone in his house for Monty. I remembered his ache on
Joan Merrow's account, our later talk with Monty, his nascent and
grim resolve that in the teeth of all the world the accident theory
should be maintained, his dismay when he had realized what a post-
mortem examination might disclose. I ran over again his whole day,
from that merry breakfast-party to the appearance of Inspector
Webster in our midst at ten o'clock at night.... Well, one peril was
now safely past. In the absence of the Exhumation Order of which
Mackwith spoke, there remained no tittle of material evidence save a
battered bit of nickel-steel in Westbury's possession or in that of the
police. If only for Esdaile's sake, I felt as if a weight had been lifted
from me.
And, on the top of all, Billy Mackwith's innocent complicity must have
moved me to that inane outburst.
"Well, for one thing, Esdaile isn't out of town at all," I said, wiping
my eyes.
"Well, I thought he was. That isn't the joke, is it?"
"Not altogether. You see——"
"Do you mean that stupid Roundabout thing?"
"No.... I beg your pardon, Billy. It came over me all of a sudden.
Now I think I can tell you——"
And so another was added to our nicely-lengthening list of Principals
in the Case.
 IV
While I spoke Billy had risen, and was pretending to examine the
prints on my walls. I continued to talk; talking was, in fact, my
morning's work that day. I finished, and there was a long silence. I
thought my barrister-friend would never have done looking at those
prints.
Then suddenly he crossed over to my table and stood leaning lightly
on his fingertips.
"Why wasn't I told this sooner?" he asked, his eyes brightly on mine.
For a moment I thought he meant that our neglect to inform him
had landed him into this equivocal position with regard to the
coroner's jury, and was beginning to explain that, being everybody's
business, it had also apparently been nobody's. But he cut me short.
"Oh, I don't mean that. Leave the inquest out of it for the present.
What I mean is that I could have saved our friend a good deal of
mental pain if I'd known—and you too," he added, "from the way
you laughed just now."
"How?" I asked.
"In this way," he replied, sitting down on the edge of my table and
giving his striped cashmere trousers a little hitch. "Say that a shot
has been fired.... Philip, I take it, has been worrying about the
consequences to this fellow Smith, and incidentally to Miss Merrow.
Now if I'd been there to ask Rooke a few material questions I think I
could have assured him that it's a thousand to one there won't be
any consequences."
"Why not?"
"The state of the body," he replied promptly. "Rooke saw it, you say,
or at any rate quite enough of it; I saw it too; and, shot or no shot,
it wouldn't have taken me two minutes to get out of Rooke that
there was no earthly possibility of proving that a shot caused death."
"You mean there were so many other good reasons?"
"Well, I'm not a doctor, but I should say at least a dozen. No wonder
that fellow Westbury—ah, that's his name, not Westcott—had to
make a bolt for it. Unless somebody can be produced who actually
saw the shot fired there won't be the ghost of a Case, and I'm
inclined to think that even then it would reduce itself to shooting
with intent to kill or wound—which is a felony, of course, but not
quite the same thing as murder. No, I think you can take it from me
that there won't be any consequences."
I pondered this for a moment. Then I saw the flaw in it. Every man
to his trade. Here was the advocate speaking, his whole acute mind
trained to one single end—the getting of his man off. But I myself
work in a different material and saw the Case from my own angle.
"One moment," I interposed. "When you say consequences you
mean legal consequences? In other words he'd slip through your
fingers simply because nobody actually saw him do it?"
"He wouldn't even be charged. That was practically a certainty
before the inquest. It's overwhelming now the other fellow's buried."
"But legal consequences are not the only kind of consequences there
are in the world."
"Oh, I'm not speaking of moral consequences. They're quite another
matter," quoth Billy.
"Not as regards Esdaile's having a rotten time over this," I differed.
"Let's look at it from another point of view for a moment. Neither
you nor I know Smith. But Hubbard and Esdaile do, and there's this
friendship between them. And mark you, friendship too isn't always
the same thing it was before the War. There were lots of men we
called friends then very much as a matter of habit; I mean it didn't
often occur to you to ask what kind of a man your friend would be
when it came to the pinch. We've all made new friends, and there
are some of the old ones whose names we never want to hear
again. You see what I mean? I mean the bond must be pretty strong
for two men like Esdaile and Hubbard to take instantly to the
thought of shielding Smith like ducks taking to water. I watched
them—it was really exciting—you could read both their faces like
books. Very well. Up to this point we're both talking the same
language. When we say consequences we mean legal consequences.
"But here's where the difference comes in. I don't know what
Hubbard's views are, as I haven't seen him since that night; but I do
know what Esdaile's are. He's shielding Smith—but only till he hears
what he's got to say for himself. He doesn't want to condemn him
unheard. I admit that in the meantime he's taken certain rather risky
steps, and my own opinion is that he won't find it very easy if he
wants to retrace them again; the river'd have to be dredged for a
pistol, for example, and Lord only knows what sort of a reason he'd
give for even having interfered at all. But my point is that he's done
nothing final yet. Smith's got to satisfy him, and if he can't—well, it
can't be quite the same between them again after that, can it?"
"You mean their friendship's broken?"
"Well, that's not quite the way I should put it. It might break, or
possibly it might not. What I mean is that a friendship with a man
who's killed other men in battle isn't the same thing as one with a
man who murders another in peace-time. It may be as good for all I
know, as I haven't done either, but obviously it isn't the same."
"No, I suppose not," Mackwith agreed. "And as for retracing his
steps, I agree with you that the best thing he can do is to keep his
mouth shut. I certainly intend to about that inquest. Life's too short
to go moving for Exhumation Orders."
"Well, next there's Joan Merrow. Exactly the same thing applies to
her. Is she going to marry a soldier or an assassin? Is Esdaile going
to let her? He's her guardian for all practical purposes, and he's got
that question to answer."
The barrister laughed. "I don't think he need worry about that. Miss
Merrow strikes me as a young woman who won't stand any
nonsense from guardians. Well," he took up his hat and stick, "I
must be getting along. I didn't expect all this when I came in, but it
seems to me the Case is over now. Barring these moral
consequences of yours, it practically ended when I gave in our
verdict yesterday."
"I hope you're right," I replied. "I thought so myself a few days ago,
though, and that evening a Police Inspector marched in."
He stopped at the door and spoke over his shoulder.
"Oh? You seem doubtful. Any reason?"
"None," I replied. "Only what women call a sort of feeling about it."
Mackwith laughed.
"We'll see about that when it comes," he said. "So long——"
 V
The surface indications were of course of the very slightest. So far
they consisted merely of the photograph in the Roundabout, my
speculations whether Hodgson had anything up his sleeve, and
similar trifles. But others were pending. The danger of the coroner's
inquest might be safely past, but at least half a dozen other rocks
loomed immediately ahead. The Aiglon Company, for example,
would want to know what had gone wrong with their machine, and
the manufacturers would be even more interested. The same with
the parachute people. The Aero Club and the Royal Aeronautical
Society both have their Accidents Investigation Committees, and it
was quite likely that claims against various Insurance Offices had
already been lodged. You cannot thrust a finger into the close web
of modern life without stirring up all manner of complexities. I
suppose it was these that I had already begun to fear.
Perhaps most immediate of all was the question of unauthorized
flying. What had that machine been doing over London at all?
Military machines come and go under orders, but not commercial
planes piloted by civilian aeronauts. Setting such things as Murder
and Manslaughter entirely on one side, was it not probable that
Smith would be called to book on this account? From our point of
view it was obviously most undesirable that he should be brought
into Court on any charge at all; but what if we couldn't prevent it?
What if, in the inhuman collision of powerful business interests
behind, the lawyers were to get to work—an Insurance Company
resist a claim, say, or the Aiglon people proceed against the
manufacturers on a point of warranty? You may think I was seeing
lions in the path, but it is never safe to reckon on meeting nothing
more formidable than a sheep. And I have nothing against lawyers
as a class. I don't think Billy Mackwith would pick my pocket of a
single sixpence. But I do believe they are like the road-mender with
the stone. He hit it with his hammer ten times without breaking it,
and was then asked whether he did not think he would have a better
chance of splitting it if he turned it splitting-edge uppermost. "How
do you know I want to split it?" he replied. I suspect even Billy of
not wanting to split it sometimes.
"Willett!" I called as I entered my office after lunch that day. "Just
get me out the latest thing about Air Navigation, will you?"
"Is the new one out yet?" Willett replied, walking to the big glass-
fronted cupboard where we keep the current papers of this kind; the
others are on the library shelves downstairs. "I think they were
withdrawn. I seem to remember sending to the Stationery Office and
being told there'd been a muddle of some sort."
"There must be something in force," I said. "I want that, whatever it
is."
"Just a moment—ah, here we are!"
Willett was both right and wrong. A certain issue of Statutory Rules
and Orders had as a matter of fact been withdrawn, but an amended
reprint was now available. He handed the slender white booklet to
me. It was dated April 30th, 1919, and had therefore been in force
for some days at the time of the Lennox Street accident.
I walked to my desk and settled down to the study of it.
I don't know that I was very much the wiser for my efforts. So much
seemed to be in the air in every sense of the word. The paper was
not even an Act, but an Order, and it seemed to me that its phrases
about "contravention of these Regulations" might in practice mean
almost anything. What, for example, did "stress of weather or other
unavoidable cause" mean? What would happen in case of a kind of
accident expressly excluded from the Order—"within a circle of a
radius of one mile from the center of a licensed aerodrome"? What
about the special cases permitted "by direction of the Secretary of
State on the recommendation of a Government Department"? I don't
mean that the intention of it all wasn't plain enough. The drafters of
the Regulations had done the best they could in a new and totally
Welcome to Our Bookstore - The Ultimate Destination for Book Lovers
Are you passionate about testbank and eager to explore new worlds of
knowledge? At our website, we offer a vast collection of books that
cater to every interest and age group. From classic literature to
specialized publications, self-help books, and children’s stories, we
have it all! Each book is a gateway to new adventures, helping you
expand your knowledge and nourish your soul
Experience Convenient and Enjoyable Book Shopping Our website is more
than just an online bookstore—it’s a bridge connecting readers to the
timeless values of culture and wisdom. With a sleek and user-friendly
interface and a smart search system, you can find your favorite books
quickly and easily. Enjoy special promotions, fast home delivery, and
a seamless shopping experience that saves you time and enhances your
love for reading.
Let us accompany you on the journey of exploring knowledge and
personal growth!

ebooksecure.com