(FREE PDF Sample) (Ebook PDF) Introductory Econometrics: Asia-Pacific 2nd Edition Ebooks

Full download ebook at ebooksecure.
com
(eBook PDF) Introductory Econometrics: Asia-

Pacific 2nd Edition
https://ebooksecure.com/product/ebook-pdf-introductory-
econometrics-asia-pacific-2nd-edition/
OR CLICK BUTTON
DOWLOAD EBOOK
Download more ebook from https://ebooksecure.com

More products digital (pdf, epub, mobi) instant
download maybe you interests ...
(eBook PDF) Introductory Econometrics Asia Pacific

Edition
http://ebooksecure.com/product/ebook-pdf-introductory-
econometrics-asia-pacific-edition/
(eBook PDF) ACCT2 Managerial 2nd Asia Pacific Edition
http://ebooksecure.com/product/ebook-pdf-acct2-managerial-2nd-
asia-pacific-edition/
(eBook PDF) Managerial Accounting Asia-Pacific Edition

2nd Edition
http://ebooksecure.com/product/ebook-pdf-managerial-accounting-
asia-pacific-edition-2nd-edition/
(eBook PDF) ACCT2 Financial 2nd Asia-Pacific Edition by

Jonathan
http://ebooksecure.com/product/ebook-pdf-acct2-financial-2nd-
asia-pacific-edition-by-jonathan/
(eBook PDF) Introduction to Corporate Finance 2nd Asia-
Pacific Edition
http://ebooksecure.com/product/ebook-pdf-introduction-to-
corporate-finance-2nd-asia-pacific-edition/
(eBook PDF) Marketing Principles 2nd Asia Pacific

Edition by William
http://ebooksecure.com/product/ebook-pdf-marketing-
principles-2nd-asia-pacific-edition-by-william/
(eBook PDF) Principles of Anatomy and Physiology 2nd

Asia-Pacific Edition
http://ebooksecure.com/product/ebook-pdf-principles-of-anatomy-
and-physiology-2nd-asia-pacific-edition/
(eBook PDF) Managerial Accounting 1st Asia Pacific

Edition
http://ebooksecure.com/product/ebook-pdf-managerial-
accounting-1st-asia-pacific-edition/
(Original PDF) Marketing Research 3rd Asia-Pacific

edition
http://ebooksecure.com/product/original-pdf-marketing-
research-3rd-asia-pacific-edition/
Copyright 2021 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part. WCN 02-200-202
contents vii
» Quadratic functions 35
» The natural logarithm 37
» The exponential function 40
→ Summary 41
→ Key terms 41
→ Review questions 41
→ Multiple choice questions 43
Chapter 03 fundamentals of statIstIcs: a revIew 46

3.1 Random variables and their probability distributions 46
» Discrete random variables 47
» Continuous random variables 48
3.2 Joint distributions, conditional distributions and independence 50
» Joint distributions and independence 50
» Conditional distributions 51
3.3 Features of probability distributions 52
» A measure of central tendency: the expected value 53
» Properties of expected values 54
» Another measure of central tendency: the median 55
» Measures of variability: variance and standard deviation 55
» Standardising a random variable 57
» Skewness and kurtosis 58
3.4 Features of joint and conditional distributions 58
» Measures of association: covariance and correlation 58
» Variance of sums of random variables 60
» Conditional expectation 61
» Properties of conditional expectation 63
» Conditional variance 64
3.5 The normal and related distributions 64
» The normal distribution 64
» The standard normal distribution 65
» The chi-square distribution 67
» The t distribution 68
» The F distribution 68
3.6 Populations, parameters and random sampling 70
» Sampling 70
3.7 Properties of estimators 71
» Estimators and estimates 71
» Unbiasedness 72
» The sampling variance of estimators 74
» Efficiency 75
» Consistency 76
» Asymptotic normality 78
0 Wooldridge1e_50839_txt_3pp.indd 7 16/09/20 10:22 AM

viii contents
3.8 Confidence intervals and hypothesis testing 79

» The nature of interval estimation 79
» Confidence intervals for the mean from a normally
distributed population 81
3.9 Hypothesis testing 83
» Testing hypotheses about the mean
in a normal population 85
» Computing and using p-values 88
» The relationship between confidence intervals
and hypothesis testing 91
→ Summary 92
→ Key terms 92
Part
2
regressIon analysIs wIth cross-sectIonal data 99
Chapter 04 the sImple regressIon model 100
4.1 Definition of the simple regression model 100
4.2 Deriving the ordinary least squares estimates 105
4.3 Examples of simple regression obtained using real data 108
» A note on terminology 111
4.4 Properties of OLS on any sample of data 112
» Fitted values and residuals 112
» Algebraic properties of OLS statistics 113
» Goodness-of-fit 115
4.5 Units of measurement and functional form 116
» The effects of changing units of measurement on OLS statistics 116
» Incorporating non-linearities in simple regression 118
» The meaning of ‘linear’ regression 120
4.6 Unbiasedness, consistency and variances of the OLS estimators 121
» Unbiasedness of OLS 121
» Variances of the OLS estimators 125
» Estimating the error variance 129
» Consistency of OLS 131
4.7 Regression on a binary explanatory variable 133
» Counterfactual outcomes, causality and policy analysis 135
→ Summary 139
→ Key terms 139
→ Computer questions 146

contents ix
Chapter 05 multIple regressIon analysIs: estImatIon 150

5.1 Motivation for multiple regression 150
» The model with two independent variables 150
» The model with k independent variables 153
5.2 Mechanics and interpretation of ordinary least squares 154
» Obtaining the OLS estimators 154
» Interpreting the OLS regression equation 156
» On the meaning of ‘holding other factors fixed’
in multiple regression 157
» Changing more than one independent variable simultaneously 158
» OLS fitted values and residuals 158
» A ‘partialling out’ interpretation of multiple regression 158
» Goodness-of-fit 159
5.3 Properties of the OLS estimators 162
» Unbiasedness of the OLS estimators 165
» The variance of the OLS estimators 166
» The components of the OLS variances: multicollinearity 167
» Estimating σ2 standard errors of the OLS estimators 169
» Efficiency of OLS: the Gauss-Markov theorem 170
» Consistency 171
5.4 Some comments on the language of multiple regression analysis 172
5.5 Several scenarios for applying multiple regression 173
» Prediction 173
» Efficient markets 174
» Measuring the trade-off between two variables 175
» Testing for ceteris paribus group differences 175
» Potential outcomes, treatment effects, and policy analysis 176
→ Summary 178
→ Key terms 179
→ 5A.1 Derivation of the first order conditions in
Equation (5.13) 192
→ 5A.2 Derivation of Equation (5.22) 192
→ 5A.3 Proof of Theorem 5.1 192
Chapter 06 multIple regressIon analysIs: Inference 195

6.1 Sampling distributions of the OLS estimators 195
6.2 Testing hypotheses about a single population parameter: the t test 198
» Testing against one-sided alternatives 200
» Two-sided alternatives 203

x contents
» Testing other hypotheses about βj 205

» Computing p-values for t tests 206
» A reminder on the language of classical hypothesis testing 208
» Economic, or practical, versus statistical significance 209
6.3 Confidence intervals 211
6.4 Testing hypotheses about a single linear combination of the
parameters 213
6.5 Revisiting causal effects and policy analysis 215
6.6 Testing multiple linear restrictions: the F test 216
» Testing exclusion restrictions 216
» Relationship between F and t statistics 221
» The R-squared form of the F statistic 221
» Computing p-values for F tests 223
» The F statistic for overall significance of a regression 223
» Testing general linear restrictions 224
6.7 Confidence intervals for predictions 226
6.8 Reporting regression results 229
→ Summary 231
→ Key terms 233
Chapter 07 model specIfIcatIon 244

7.1 More on functional form 244
» More on using logarithmic functional forms 244
» Models with quadratics 248
» Models with interaction terms 253
7.2 Specification errors 254
» Including irrelevant variables in a regression model 255
» Omitted variable bias: the simple case 255
» Omitted variable bias: more general cases 258
» Variances in misspecified models 259
» Functional form misspecification 260
» RESET as a general test for functional form misspecification 262
» Tests against non-nested alternatives 263
7.3 Multicollinearity 264
→ Summary 268
→ Key terms 269

contents xi
Chapter 08 multIple regressIon analysIs wIth qualItatIve

InformatIon: BInary (or dummy) varIaBles 280
8.1 Describing qualitative information 280
8.2 A single dummy independent variable 281
» Using a dummy independent variable to account for outliers 285
» Interpreting coefficients on dummy explanatory variables when the
dependent variable is log(y) 286
8.3 Using dummy variables for multiple categories 288
» Incorporating ordinal information by
using dummy variables 289
8.4 Interactions involving dummy variables 292
» Interactions among dummy variables 292
» Allowing for different slopes 293
» Testing for differences in regression functions across groups 295
8.5 A binary dependent variable: the linear probability model 298
8.6 More on policy analysis and program evaluation 301
» Program evaluation and unrestricted regression adjustment 303
8.7 Interpreting regression results with discrete dependent variables 306
→ Summary 307
→ Key terms 308
Chapter 09 heteroscedastIcIty 319

9.1 Consequences of heteroscedasticity for OLS 319
9.2 Heteroscedasticity-robust inference after OLS estimation 320
9.3 Testing for heteroscedasticity 324
» The Breusch-Pagan test for heteroscedasticity 326
» The White test for heteroscedasticity 327
» A special case of the White test for heteroscedasticity 328
9.4 Weighted least squares estimation 328
» The heteroscedasticity is known up to a multiplicative constant 329
» The heteroscedasticity function must be estimated: feasible GLS 333
» A feasible GLS procedure to correct for heteroscedasticity 334
» What if the assumed heteroscedasticity function is wrong? 336
9.5 The linear probability model revisited 338
» Estimating the linear probability model by weighted least squares 339
→ Summary 341
→ Key terms 341

xii contents
Part regressIon analysIs wIth tIme serIes data 349
3
Chapter 10 BasIc regressIon analysIs wIth tIme serIes data 350
10.1 The nature of time series data 350
10.2 Examples of time series regression models 351
» Static models 352
» Finite distributed lag models 352
» A convention about the time index 354
10.3 Finite sample properties of OLS under classical assumptions 355
» Unbiasedness of OLS 355
» The variances of the OLS estimators and the Gauss-Markov theorem 358
» Inference under the classical linear model (CLM) assumptions 360
10.4 Functional form, dummy variables, index numbers and event studies 362
» Functional forms 362
» Dummy variables 362
» Index numbers 363
» Event study 365
10.5 Trends and seasonality 367
» Characterising trending time series 367
» Using trending variables in regression analysis 369
» A detrending interpretation of regressions with a time trend 371
» Computing R-squared when the dependent variable is trending 372
» Seasonality 374
10.6 Stationary and weakly dependent time series 376
» Stationary and non-stationary time series 376
» Weakly dependent time series 378
10.7 Asymptotic properties of OLS 380
→ Summary 385
→ Key terms 387
Chapter 11 serIal correlatIon and heteroscedastIcIty

In tIme serIes regressIons 394
11.1 Properties of OLS with serially correlated errors 394
» Unbiasedness and consistency 394
» Efficiency, inference and goodness-of-fit 395
» Serial correlation in the presence of lagged dependent variables 396
11.2 Testing for serial correlation 397
» A t test for first order serial correlation with strictly exogenous regressors 397
» The Durbin-Watson test under classical assumptions 399
» Testing for first order serial correlation without strictly exogenous
regressors 400
» Testing for higher order serial correlation 402

contents xiii
11.3 Correcting for serial correlation with strictly exogenous regressors 404
» Obtaining the best linear unbiased estimator in the AR(1) model 404
» Feasible GLS estimation with AR(1) errors 406
» Comparing OLS and FGLS 407
» Correcting for higher order serial correlation 408
11.4 Serial correlation-robust inference after OLS 409
» Serial correlation-robust standard error for β̂ 1 411
11.5 Heteroscedasticity in time series regressions 413
» Heteroscedasticity-robust statistics 413
» Testing for heteroscedasticity 414
» Autoregressive conditional heteroscedasticity 414
» Heteroscedasticity and serial correlation in regression models 416
→ Summary 417
→ Key terms 418
Chapter 12 modellIng hIghly persIstent tIme serIes data 424

12.1 Highly persistent time series 424
» Transformations on highly persistent time series 428
12.2 Testing for unit roots 429
12.3 Spurious regression 433
12.4 Cointegration and error correction models 435
» Cointegration 435
» Error correction models 438
→ Summary 439
→ Key terms 440
Part addItIonal topIcs 445
4
Chapter 13 panel data models 446
13.1 Introduction to panel data 446
13.2 Panel data model: error components model 448
» Exogeneity assumption for the panel data model 448
» Pooled least squares 449

xiv contents
13.3 Fixed effects estimator 449

» Difference-in-difference estimator 450
» Fixed effects estimator 451
» The least squares dummy variable (LSDV) estimator 455
13.4 Random effects estimator 456
» Random effects or fixed effects? 460
» Fixed and random effects with unbalanced panels 460
→ Summary 461
→ Key terms 461
→ 13A.1 Assumptions for fixed and random effects 467
→ 13A.2 Inference robust to serial correlation and heteroscedasticity
for fixed effects and random effects 470
Chapter 14 lImIted dependent varIaBle models

and sample selectIon correctIons 471
14.1 Logit and probit models for binary response 472
» Logit and probit models: specifications 472
» Maximum likelihood estimation of logit and probit models 475
» Testing multiple hypotheses 475
» Interpreting the logit and probit estimates 476
14.2 The Tobit model for corner solution responses 483
» Interpreting the Tobit estimates 484
» Specification issues in Tobit models 489
14.3 The Poisson regression model 490
14.4 Censored and truncated regression models 494
» Censored regression models 495
» Truncated regression models 497
14.5 Sample selection corrections 499
» When is OLS on the selected sample consistent? 500
» Incidental truncation 501
» Sample selection correction 503
→ Summary 505
→ Key terms 506
→ Computer exercises 510
→ 14A.1 Maximum likelihood estimation with explanatory variables 514
→ 14B.1 Asymptotic standard errors in limited dependent variable
models 515

contents xv
Chapter 15 an IntroductIon to machIne learnIng

In econometrIcs 516
15.1 Machine learning: an introduction 516
15.2 Penalised linear models 518
» The linear regression model 518
» The ridge regression model 519
» The LASSO regression model 519
» The elastic net regression model 520
15.3 Forecasting and time series cross-validation 522
» Types of regression models used for forecasting 524
» One-step-ahead forecasting 524
» Comparing one-step-ahead forecasts 526
» Multiple-step-ahead forecasts 528
» Time series cross-validation 529
→ Summary 530
→ Key terms 531
APPENDIX A Statistical tables 539

Glossary 551
References 563
Index 565

PrefaCe
As with the previous edition, the emphasis in assumption of non-random explanatory variables.
this revised Asia–Pacific edition is on applying As a non-trivial benefit, the population model/
econometrics to real-world problems. The focus is random sampling approach reduces the number
on understanding and interpreting the assumptions of assumptions that students must absorb and
in light of actual empirical applications. It has been understand.
designed primarily for undergraduate students The focus here on the population model
undertaking their first study of econometrics. The emphasises that the fundamental assumptions
concise structure and simplified explanations provide underlying regression analysis, such as the zero
a clear introduction to the subject and bridge students’ mean assumption on the unobservable error term,
transition from basic statistics into econometrics. are properly stated conditional on the explanatory
variables. This leads to a clear understanding of the
Organised fOr tOday’s kinds of problems, such as heteroscedasticity (non-
ecOnOmetrics instructOr constant variance), that can invalidate standard
The most noticeable feature that distinguishes this text inference procedures.
from most others is the separation of topics by the kind Because the assumptions for cross-sectional
of data being analysed. This is a clear departure from analysis are relatively straightforward yet realistic,
the traditional approach, which presents a linear model, students can get involved early with serious cross-
lists all assumptions that may be needed at some future sectional applications without having to worry about
point in the analysis, and then proves or asserts results other issues of trends, seasonality, serial correlation,
without clearly connecting them to the assumptions. high persistence and spurious regression that are
It has been divided into four parts. Part 1 of the text ubiquitous in time series regression models. These
covers key concepts from mathematics and statistics topics are covered in Part 3 of the text. By postponing the
required for econometric analysis. This provides readers econometric analysis of time series data, proper focus
with the background mathematics and basic probability can be placed on the potential pitfalls in analysing time
and statistics required for undertaking the text. It also series data that do not arise with cross-sectional data. In
provides a brief review of this material for those readers effect, time series econometrics finally gets the serious
that have a prior background in these areas. Wherever treatment it deserves in an introductory text. Students’
necessary, explanations and details of derivations are learning experience with regard to the basic time series
provided with regard to the theorems and formulas. econometrics is supplemented by some additional
Part 2 examines the multiple regression analysis with special interest topics introduced in Part 4. Building an
cross-sectional data, under the assumption of random elementary knowledge base on pooled and panel data
sampling. This setting is natural to students because they methods, limited dependent variable models and big
are familiar with random sampling from a population data with machine learning should appropriately orient
in their introductory statistics courses. Importantly, it the students to further their studies in econometrics to
allows us to distinguish assumptions made about the suit the analytical skill requirements of the contemporary
underlying population regression model – assumptions business world. The machine learning chapter presents
that can be given economic or behavioural content – a clear and succinct introduction to tools that are now
from assumptions about how the data were sampled. indispensable to analyse business data sets with a large
Discussions about the consequences of non-random number of observations and variables.
sampling can be treated in an intuitive fashion after the Further, each topic is presented in a systematic
students have a good grasp of the multiple regression way by building on the previous material in a logical
model estimated using random samples. fashion, and assumptions are introduced only as
An important feature of a modern approach is that they are needed to obtain a conclusion. For example,
the explanatory variables – along with the dependent empirical researchers who use econometrics in their
variable – are treated as outcomes of random variables. research understand that not all of the Gauss-Markov
For the social sciences, allowing random explanatory assumptions are needed to show that the ordinary least
variables is much more realistic than the traditional squares (OLS) estimators are unbiased. Yet the vast
xvi Copyright 2021 Cengage Learning. All Rights Reserved. May not be copied, scanned, or duplicated, in whole or in part. WCN 02-200-202

preface xvii
majority of econometrics texts introduce a complete The AsiA–PAcific ediTion

set of assumptions (many of which are redundant or in
As with the previous Asia–Pacific edition, this revised
some cases even logically conflicting) before proving
edition includes a number of distinct features. To
the unbiasedness of OLS. Similarly, the normality
begin with, the target audience is undergraduate
assumption is often included among the assumptions
students undertaking their first study of econometrics.
that are needed for the Gauss-Markov theorem, even
Chapters of background material on introductory
though it is fairly well known that normality plays no
mathematics, probability and statistics have been
role in showing that the OLS estimators are the best
included to provide a background for those students
linear unbiased estimators.
who have not previously been introduced to this
This systematic approach is illustrated by the
material or a brief review for those who have. The body
order of assumptions used for multiple regression in
of the book emphasises cross-section and time series
Part 2. This structure results in a natural progression
applications as well as providing an introduction to
for briefly summarising the role of each assumption:
more advanced material including an introduction to
■ MLR.1: Introduce the population model and panel data models, limited dependent variable models
interpret the population parameters (which we and machine learning applications.
hope to estimate). Throughout this edition many local examples and
■ MLR.2: Introduce random sampling from the data sets are used that provide a practical approach
population and describe the data that we use to to teaching this highly mathematical subject and
estimate the population parameters. demonstrate the application of econometrics to
■ MLR.3: Add the assumption on the explanatory real-world problems that students can relate to. The
variables that allows us to compute the estimates end-of-chapter problems and computer exercises
from our sample; this is the so-called no perfect are heavily oriented toward empirical work, rather
collinearity assumption. than complicated derivations, and students are
■ MLR.4: Assume that, in the population, the mean asked to reason carefully based on what they have
of the unobservable error does not depend on learned. A number of the exercises also allow the
the values of the explanatory variables; this is the student to examine and replicate econometric
‘mean independence’ assumption combined with analyses drawn from recently published research.
a zero population mean for the error, and it is the In addition, a set of multiple choice questions are
key assumption that delivers unbiasedness of OLS. included in every chapter.
After introducing Assumptions MLR.1 to MLR.3, The introductory chapters of this new edition come
one can discuss the algebraic properties of ordinary with additional explanations of some basic concepts
least squares – that is, the properties of OLS for a as well as updates of graphs and relevant data. Chapter
particular set of data. By adding Assumption MLR.4, 1 includes a section on the importance of computer-
we can show that OLS is unbiased (and consistent). generated graphs. These visualisation techniques
Assumption MLR.5 (homoscedasticity) is added for the have become important tools in the preliminary
Gauss-Markov theorem and for the usual OLS variance analysis of data. Such detail is rare in an introductory
formulas to be valid. Assumption MLR.6 (normality), econometrics textbook. Models and inferences are
which is not introduced until Chapter 6, is added to presented with increased clarity. To aid students’
round out the classical linear model assumptions. The understanding at the very outset, Chapter 2 expands
six assumptions are used to obtain exact statistical discussions on non-linear functions and Chapter 3
inference and to conclude that the OLS estimators have incorporates a fresh explanation on the difference
the smallest variances among all unbiased estimators. between economic and statistical significance.
Throughout the text, ceteris paribus relationships There is an extensive glossary at the end of the
are emphasised, which is why, after one chapter on book. The short definitions and descriptions are a
the simple regression model, we move to multiple helpful refresher for students studying for exams or
regression analysis. The multiple regression reading empirical research that uses econometric
setting motivates students to think about serious methods.
applications early. Prominence is also given to policy
analysis; practical topics such as interpreting partial
effects in models with interaction terms, are covered
in a simple fashion.
0 Wooldridge1e_50839_txt_3pp.indd 17 24/09/20 5:01 PM

Guide to the text
As you read this text you will find a number of features in every
chapter that will enhance your study of econometrics and help you
to understand how the theory is applied in the real world.
Part-oPening features
Part openers introduce each
1
Part of the chapters within the
part and give an overview of
how they relate to each other.
IntroductIon and
revIew
Chapter 1 The nature of econometrics and economic data
Chapter 2 Basic mathematical tools
Chapter 3 Fundamentals of statistics: a review
P
art 1 of the text covers key concepts from mathematics and statistics required for
econometric analysis. This provides the background mathematics and basic probability
and statistics required for undertaking the text. This also provides a brief review of this
material for those readers that have a prior background in these areas. Chapter 1 discusses the
scope of econometrics and raises general issues that arise in the application of econometric
methods. Chapter 2 covers some basic mathematics that are used in econometric analysis,
while Chapter 3 reviews the statistical ideas that are used at various stages in the text. Wherever
necessary, explanations and details of derivations are provided with regard to the theorems and
formulas.
ChaPter-oPening features
ChaPter 1
the nature of econometrIcs Identify the key concepts you

and economIc data will engage with through the
neW Learning objectives at
Learning objectives Overview
1.1 Understand the nature and scope of Chapter 1 discusses the scope of econometrics and
1
the start of each chapter.
econometrics. raises general issues that arise in the application
1.2 Understand the structure of empirical economic of econometric methods. Section 1.1 provides a
analysis based on notion of economic models brief discussion about the purpose and scope of
The Chapter overview

and econometric model specifications. econometrics, and how it fits into economics analysis.
Section 1.2 provides examples of how one can start
1.3 Develop knowledge on various types of data
introduces the topics that are
with an economic theory and build a model that
used in econometric analysis.
can be estimated using data. Section 1.3 examines
1.4 Analyse data by plotting them using various
graphical methods.
the kinds of data sets that are used in business,
economics and other social sciences. In Section 1.4 covered in the chapter.
1.5 Evaluate and establish meaningful relationships we present some techniques for presenting plots of
of the variables using the framework of data. Section 1.5 provides an intuitive discussion of the
econometric analysis. difficulties associated with the inference of causality in
1.6 Understand the content of the topics covered the social sciences. Section 1.6 provides an outline of
in various parts of the book. the topics covered in this book.
1.1 what Is econometrIcs?

Imagine that you are hired by your state government to evaluate the effectiveness of a publicly
Copyright 2021funded job training
Cengage program.
Learning. AllSuppose
RightsthisReserved.
program teaches
Mayworkers
not bevarious waysscanned,
copied, to use or duplicated, in whole or in part. WCN 02-200-202
computers in the manufacturing process. The 20-week program offers courses during non-
working hours. Any hourly manufacturing worker may participate, and enrolment in all or
part of the program is voluntary. You are to determine what, if any, effect the training program
has on each worker’s subsequent hourly wage.
Now, suppose you work for an investment bank. You are to study the returns on different
investment strategies involving short-term Australian treasury notes to decide whether they
comply with implied economic theories.
The task of answering such questions may seem daunting at first. At this point, you may only
dependent variable.
Chapter 5 » M u lt i p l e r e g r e s s i o n a n a ly s i s : e s t i M at i o n 165
examPle EGM expenditures and number of EGMs
In the electronic gaming machine expenditures example in (4.23), R2 = .366. Therefore, the
4.10 cause Assumption MLR.3 to fail, but it can be considered very unlikely unless we have an
number of EGMs per 1000 adults explains around 37% of the variation in electronic gaming
extremely small sample size.
machine expenditure per adult.
The final, and most important, assumption needed for unbiasedness is a direct extension
of Assumption SLR.4.
mlr.4
GUIDE TO THE TEXT xix
↪
Zero conditional mean
4.5 units of
assumption measuRement
The error u has an expected value ofand functional
zero given foRm
any values of the independent variables. In
other words,
Two important issues in applied economics are (1) understanding how changing the units
E(u|x1, x2, . . . , xk) = 0. [5.36]
of measurement of the dependent and/or independent variables affects OLS estimates, and
(2) knowing how to incorporate popular functional forms used in economics into regression
features Within ChaPters 4
One way that Assumption MLR.4 can fail is if the functional relationship between the
analysis. The mathematics needed for a full understanding of functional form issues is
Part 1 » i n T r o d uexplained
cTion and r and
e v i eexplanatory
w variables is misspecified in Equation (5.31): for example, if we
reviewed in Chapter 2.
forget to include the quadratic term inc2 in the consumption function cons = b0 + b1inc +
b2inc2 + u when we estimate the model. Another functional form misspecification occurs
Analyse practical applications of examPle
the
when eFFects
we use the level oFofchAnging a variable when the units
log of theoFvariable meAsurement
is what actually shows up in
on
the ols stAtistics
Economic
population
model of crime
model, or vice versa. These types of misspecifications and their consequences
concepts with the example boxes. 1.1 In
In a seminal article, Nobel Prize winner Gary Becker postulated a utility maximisation framework
willExample
be discussed
to describe 4.3, we chose
in Chapter
an individual’s to 7.measure
participation annual salary in thousands
in crime. Certain crimes have clear of dollars,
economic and rewards,
the return
onbut equity
mostwas
Assumption criminal measured
MLR.4
behaviours canasalsoahave
percentage
failcosts. (rather
if an important
The than as
factor
opportunity thata isdecimal).
costs correlated
of crime preventIt with
is crucial
any to know
1 , x 2 , , x k
the xcriminal
how
isfrom salary
omitted. Withroe
and
participating areother
multiple
in measured
regression
activitiesin this example
analysis,
such we are
as legal in order
able totoinclude
employment. make sense
In addition,manyof the are
factors
there estimates
among
costs the in
associated
Equation
explanatory with
(4.21). the possibility of being caught and then, if convicted,
variables, and omitted variables are less likely to be a problem in multiple regression the costs associated with
incarceration.
We must
analysis thanalso inFrom
know
simple Becker’s
that OLS
regression perspective,
analysis.the
estimates decision
change to undertake
in entirely
Nevertheless, expected
in any illegal
ways
application, activity
thereisthe
when one
are of of
units
always
resource allocation, with the benefits and costs of competing activities taken into account.
measurement
factors that, due of tothedata
dependent
limitations andorindependent
ignorance, wevariables
will not be change. Example
able toIninclude. If we4.3, thinksuppose
these
Under general assumptions, we can derive an equation describing the amount of time
that,
factors
spentrather
should
in criminalthan measuring
be controlled
activity as afor salary
and they
function inofthousands
are correlated
various of dollars,
factors. withmight
We we or
one measure
more ofsuch
represent itthein aindependent
dollars.
functionLet as
salardol
variables,bethen salary in dollars (salardol
Assumption MLR.4 will = 845 761 would
be violated. Webe will interpreted
derive thisas bias$845 761). Of7.course,
in Chapter
salardol haswea simple y = f relationship
(x1, x2, x3, x4, xto5, the x , xsalary
), measured [1.1]
in thousands of dollars: salardol =
Before show the unbiasedness 6of the 7 OLS estimators under MLR.1 to MLR.4, a word
1000
ofwhere salary.Beginning
caution. We do notstudents need to of actually run the sometimes
econometrics regression ofconfuse salardol on roe to know
Assumptions MLR.3 that and
the
y = hours
estimated
MLR.4, butequation
spent
they are in is:
criminal
quite different. activities, Assumption MLR.3 rules out certain relationships among
Consolidate your learning with the the = ‘wage’ for anorhour
x1 independent
x2 = hourly wage in legal
spent
explanatory in
1000 × salary
 employment,
criminal activity,
variables and has
 Chapter 5 »nothing
M u lt i p lto
= 963.191 + 18.501 roe
estimation
e rdo
e g rwith
e s s i othe a ly s i su.
n a nerror, : eYou
s t i Mwill
at i o know
n 165
immediately when carrying out OLS whether or not Assumption MLR.3 holds.
x3 = income other thanfrom crime or employment,
1000
numbered Key equations that are On the other
x4 = probability
cause Assumption
hand, Assumption
of getting
MLR.3 
salarydol
caught, = ( 963191
to fail,
MLR.4 –
but it can
the
be) considered
much
× 1000 + (18 501 × 1000
more
very) roe
important of the two –
unlikely unless we have an
restricts the
relationship between the convicted
unobserved factors in u and the explanatory variables. Unfortunately,
listed and referred to in the text. we
x5 = probability
extremely
x6 will
small of
never know
= expected
being
sample
for sure
sentence
size.  if caught,
salardol
whether
if convicted, the=average
and 963191 + 18 501
value the. unobserved factors is unrelated
of roe [4.36]
110 Part 2 » R e g R e s s i o n aThe
n a l yfinal,
s i s w i and
t h c Rmost
o s s - s eimportant,
c t i o n a l d a tassumption
a needed for unbiasedness is a direct extension
toxthe= age.
explanatory variables. But this is the critical assumption.
of 7Assumption SLR.4.
Other factors generally affect a person’s decision to participate in crime, but the list above
Examine theoretical concepts with the mlr.4 is representative of what might result from a formal economic analysis. Since economic
unbiasedness of The ols esTimaTors
↪
examPle Zero conditional
variables are interrelated, mean it is possible to specify a variable of interest as a function of many
assumption House price and land size
assumption and theorem boxes. 4.4
Weother
Let yare now
= the
other words,
function
assumptions. f (•)
land size of theAs
u hasAsanisto
The variables.
error
saleready
in (1.1).
expected
common
price of a show
house in
This
in the(land
property function
simple
value
in of zero given
economic
unbiasedness
Windsor, Ontario
sizeregression depends
) measured inon
case,
anywe
theory,
of OLS values
have
under
(sell) measured
an
theunderlying
square
ofthe
expectations
metres.
thefirst
not independent
been
utility
specific
in dollarsfour
function,
are
variables.
about the
and multiple
which
conditional onisthe
In
let x = the regression
rarely
values
known.
Using Nevertheless,
the data in
E(u|x we
ANGLIN.RAW
, x can
, . . . use
, xfrom
) economic
= Anglin
0. theory
and Gencay – or introspection
(1996), where n =–546
to predict the[5.36]
observations effect
of the explanatory variables 1 2 in the k sample, something we show explicitly in Appendix 5A (see
onthat each
house variable
prices sold would
during have on criminal
July, August activity. This
and September is we
1987, theobtain
basis for
the an econometric
following OLS analysis
end of chapter)
of individual but not in the text.
regression line criminal
(or sample activity.
regression function):
One way that Assumption MLR.4 can fail is if the functional relationship between the
5.1  = 34 136.19 + 71.03 land size
explained
Unbiasednessand sell of OLS variables is misspecified in Equation (5.31): for example, if we
explanatory
↖theorem forget to include
Under Assumptions = 546,
nthe quadratic
MLR.1 to MLR.4, term inc2 in the consumption function[4.22] cons = b0 + b1inc +
Formal economic modelling is sometimes the starting point for empirical analysis, but it
inc
bwhere 2
+
the u when
intercept we
andβˆestimate , j =the
= β j estimates
slope model.
have
... , kbeen Another
rounded functional
to two decimal form misspecification occurs
is more
2 common toE(use j ) economic 0, 1,theory , less formally, or even to places. We must [5.37]
rely entirely on intuition.
when
interpret wethisuseequation
the level with of caution.
a variable Thewheninterceptthecorresponds
log of the variable is what sale
to the estimated actually
price shows up in
Youwhenmay
forlandagree
any that
values
size ofthe
ismodel,
equal determinants
thetopopulation parameterofminimum
criminal
bj. In value
otherbehaviour
words,
land size appearing
the OLS in Equation
theestimators are unbiased (1.1) are
the estimators
population of the
or0.vice
population
However,
versa.the These types of
of misspecifications in sample
and their is around
consequences
reasonable
153 squarebased metres. onFor a propertyparameters.
common sense;
with a land we size
might of 153arrive
square atmetres,
such an theequation
predicted sell directly, without
will  be discussed in Chapter 7.
is sell
starting =from
34 136.19
utility+ 71.03(153)
maximisation.= 45 003.78 , or $45
This view 003.78
has (in some1987 merit,
dollars).although
From the Bank there of are cases in
Assumption
Canada MLR.4 can also fail if an important
website (http://www.bankofcanada.ca) the 1987factor thatisisequal
cpi value correlated
to 56.1 with
whenany 2013 x1 , x 2 ,, x k
which
is formal that
cpiomitted.
= 100 so With
derivations
multiple
$45
provide
003.78 regression
insights
(in 1996 dollars)
thatwe
analysis,
is equal
intuition
toare$80able can
220.64
overlook.
to include
(in 2013 many dollars).factors among the
Next The is an
slope example
estimate of
in an equation
(4.22) implies that
that we
one can
more derive
square through
metre
explanatory variables, and omitted variables are less likely to be a problem in multiple regression of somewhat
land size informal
increases reasoning
(seethe house sale1.2).
Example price by $71.03. Because of the linear nature of (4.22), another square metre of
analysis than in simple regression analysis. Nevertheless, in any application, there are always
land size increases the house sale price by the same amount, regardless of the initial land size.
iCons factors
In Section
factors
explanatory
that,
4.5,due
should
we to data limitations
discuss some methods or ignorance,
that allow for wenon-constant
will not be able
be controlled for and they are correlated with one or more of the independent
variables.
to include.
marginal effectsIfofwe ourthink these
examPle Job training and worker productivity

variables, then Assumption MLR.4 will be violated. We will derive this bias in Chapter 7.
Consider the problem posed at the beginning of Section 1.1. A labour economist would like to
Before we show the unbiasedness of the OLS estimators under MLR.1 to MLR.4, a word
Examples marked with the Continuous
1.2 examine the effects of job training on worker productivity. CHAPTERIn 4 »this
T h ecase,
s i m p lthere
e r e g ris
e slittle
s i o n need
m o d e lfor 139
examPle of caution. Beginning
Electronic
formal economic gaming students(EGM)
machine
theory.
of econometrics
Basic economic expenditures sometimes
understanding and number confuse
is sufficient of Assumptions
gaming
for realising machines MLR.3
that factors
and
MLR.4, asbut they areexperience
quitefor and Assumption
different. MLR.3
worker rules out certain relationships among
case icon highlight the running case on
The
such Victorian Commission
education, Gambling and Liquor
training affectRegulation reports
productivity. annual data
Also, for local are
economists
4.5
the
well independent
government areas
aware that workers or explanatory
(LGAs) onare paidvariables
EGM expenditures
commensurate and haswith
per adultnothing
theirto
and the do with theThis
number
productivity. of error,
EGMs u. Youreasoning
per
simple 1000 will know
gaming machines that demonstrates summAry

adults.
immediately
leads to Thea datawhen
model in the fileasPOKIES.RAW contains data for the 2012/13 fiscal year for EGM
carrying
such out OLS estimation whether or not Assumption MLR.3 holds.
expenditures per adult and the number of EGMs per 1000 adults. We estimate a simple
On
We the
have other
regression modelwage
hand,
introduced Assumption
the
to find= out simple
f (educ, MLR.4
linear
whetherexper, –
regression the
training),
increasing
much
model more
in this important
chapter, of
and the
we
the number of EGMs per 1000 adults implies [1.2]
two
have – restricts
covered the
how key concepts throughout the text its basic properties.
relationship
a higher
estimate
EGMbetween
expenditure
the slopeequation
Given
theper aunobserved
random
adult. sample, factors theinmethod
u and the ofexplanatory
ordinary least squares is
variables. used to
Unfortunately, ▻
we willThe never
estimatedknowand forintercept
sure parameters
whether
using the 70 the in the population
average
observations value of themodel.
is unobservedWe have demonstrated
factors is unrelated
work together. the algebra of the OLS regression line, including computation of fitted values and residuals,
to the explanatory
and the obtaining exp_ variables.
 But
predicted=changes
of per_adult this
148.48 +in is the
53.72 critical
EGMs
the dependent assumption.
variable for a given change in the
independent variable. In Section n = 70,4.5, we discussed two issues of practical importance: [4.23] (1) the
unbiasedness
behaviour
where the
of the OLS estimates of estimates
The
whenols esTimaTors
we change the units of measurement of the dependent
variable or intercept and slope
the independent variable, andhave been
(2) the userounded to two log
of the natural decimal places.
to allow for constant
We
This are now
estimated
elasticity ready
equation
and constant tosemi-elasticity
show
impliesunbiasedness
that if models.
the number of OLS under
of EGMs perthe1000 first fourincreases
adults multiple by regression
1 then the annual expenditure per adult willthe
increase by $53.72. OfSLR.1
In Section
assumptions. 4.6,
As in wetheshowed
simple that, under
regression case, fourthe expectationscourse,
Assumptions many other
are through
conditional SLR.4, onthetheOLS
values
end-of-ChaPter features factors willare
estimators
of the explanatory
disadvantaged
value
affect this relationship,
unbiased.
it is.
of the independent
The key including the
assumption is geographic
that the location
error term
variables in the sample, something we show explicitly in Appendix u of thezero
has
variable x. Unfortunately, there are reasons to think this is false in
LGAmeanand how given 5Aany(see
end of chapter) but not in the text.
many social science applications of simple regression, where the omitted factors in u are often
correlated with x. We have also shown that these four assumptions imply that OLS is consistent.
5.1
At the end of each chapter you will find several tools to help you to review, practise and extend your
When
↖theorem formulas
we add the assumption
Unbiasedness
Underfor the sampling
Assumptions
of OLS
MLR.1
that the variance of the error given x is constant, we get simple
variances
to MLR.4, of the OLS estimators. As we saw, the variance of the slope
knowledge of the key learning outcomes. estimator βˆ1 increases as the error variance increases, and it decreases when there is more
sample variation inE( the
ˆ
β j )independent
= β j , j = 0, 1, variable.
... , k , We alsoCHAPTERderived 4 » Tan
h eunbiased
simple reg estimator [5.37]
r e s s i o n mfor
odel 139
σ 2 = var(u ) .
for any values of the population parameter bj. In other words, the OLS estimators are unbiased
In Section 4.7 we covered the important case where x is a binary variable, and showed that
estimators of the population parameters.
the OLS ‘slope’ estimate is simply βˆ0 = y 1 − y 0, the difference in the averages of yi between the
summAry
xi = 1 and xi = 0 subsamples. We also discussed how, in the context of causal inference, βˆ1 is an
unbiased estimator of the average treatment effect under random assignment into the control
We have introduced the simple linear regression model in this chapter, and we have covered
and treatment groups. In Chapter 5 and beyond, we will study the case where the intervention
its basic properties. Given a random sample, the method of ordinary least squares is used to
or treatment is not randomised, but depends on observed and even unobserved factors.
estimate the slope and intercept parameters in the population model. We have demonstrated
Much work is left to be done. For example, we still do not know how to test hypotheses
the algebra of the OLS regression line, including computation of fitted values and residuals,
about the population parameters, 0 and 1. Therefore, although we know that OLS is unbiased
and the obtaining of predicted changes in the dependent variable for a given change in the
for the population parameters under Assumptions SLR.1 to SLR.4, we have no way of drawing
independent variable. In Section 4.5, we discussed two issues of practical importance: (1) the
inferences about the population. Other topics, such as the efficiency of OLS relative to other
behaviour of the OLS estimates when we change the units of measurement of the dependent
possible procedures, have also been omitted.
variable or the independent variable, and (2) the use of the natural log to allow for constant
The issues of confidence intervals, hypothesis testing and efficiency are central to multiple
elasticity and constant semi-elasticity models.
regression analysis as well. Since the way we construct confidence intervals and test statistics is
In Section 4.6, we showed that, under the four Assumptions SLR.1 through SLR.4, the OLS
Review your understanding of the very similar for multiple regression – and because simple regression is a special case of multiple
estimators are unbiased. The key assumption is that the error term u has zero mean given any
regression – our time is better spent moving on to multiple regression, which is much more
value of the independent variable x. Unfortunately, there are reasons to think this is false in
key chapter topics with the chapter widely applicable than simple regression. Our purpose in Chapter 4 was to get you thinking
many social science applications of simple regression, where the omitted factors in u are often
about the issues that arise in econometric analysis in a fairly simple setting.
correlated with x. We have also shown that these four assumptions imply that OLS is consistent.
summary and Key terms list. When we add the assumption that the variance of the error given x is constant, we get simple
formulas for the sampling variances of the OLS estimators. As we saw, the variance of the slope
Key terms
estimator βˆ1 increases as the error variance increases, and it decreases when there is more
sample
averagevariation
treatmentineffect the independent variable.
explained sum Weofalso derived an predictor
squares unbiasedvariable
estimator for
σ 2(ATE)
= var(u ) . (SSE) random assignment
In Section
average causal4.7 we covered
effect (ACE) theexplained case where x is a binary
importantvariable variable, and
randomised showed
control trialthat
(RCT)
the
binary OLS ‘slope’ estimate
(dummy) βˆ0 = y 1 − y 0,variable
variable is simply explanatory averages of yi between the
the difference in theregressand
i
= 1 and
xcausal xi = 0 subsamples.
(treatment) effect We also fitted discussed
value how, in the contextregressor of causal inference, βˆ1 is an
unbiased
coefficient estimator of the averagefirst
of determination treatment effect under random residual
order conditions assignment into the control
and treatment groups. In Chapterheteroscedasticity
consistency 5 and beyond, we will study the case residualwhere
sumthe of intervention
squares (SSR)
or treatment
constant is notmodel
elasticity randomised, but depends on observed and even
homoscedasticity unobserved
response variablefactors.
Much
control work is left to be done. independent
group For example,variable we still do not knowR-squaredhow to test hypotheses
about
controlthe population parameters,intercept
variable 0 and 1parameter
. Therefore, although we know regression
sample that OLS is function
unbiased
for the population parameters under
covariate OLSAssumptions
regression line SLR.1 to SLR.4, we(SRF) have no way of drawing
inferences
dependentabout variable the population. Other mean topics,
independent such as the efficiency of OLS relative to other
semi-elasticity
possible
elasticityprocedures, have also been omitted.regression
population simple linear regression
error The termissues of confidence intervals,
(disturbance) hypothesis
function (PRF) testing and efficiency modelare central to multiple
Copyright 2021 Cengage Learning. All Rights Reserved. May not be copied,
regression scanned,
error variance analysis asor well.duplicated,
Since the way in
predicted we whole
constructor
variable in part.
confidence WCN
intervals
slope and
parameter02-200-202
test statistics is
very similar for multiple regression – and because simple regression is a special case of multiple
regression – our time is better spent moving on to multiple regression, which is much more
widely applicable than simple regression. Our purpose in Chapter 4 was to get you thinking
about the issues that arise in econometric analysis in a fairly simple setting.
0 Wooldridge1e_50839_txt_3pp.indd 19 Key terms 16/09/20 10:22 AM

average treatment effect explained sum of squares predictor variable
(ATE) (SSE) random assignment
xx GUIDE TO THE TEXT
140 Part 2 » R e g R e s s i o n a n a l y s i s w i t h c R o s s - s e c t i o n a l d a t a
end-of-ChaPter
standard error of the features
total sum of squares (SST) zero conditional mean
regression (SER) treatment group assumption
sum of squared residuals (SSR)
Test your knowledge and consolidate

review questions
1 Suppose you are interested to examine the effect of women’s education on their fertility
your learning through the review
in Australia. Let C denote the number of children ever born to a woman, and let E denote
years of education for the woman. A simple model relating fertility to years of education is questions.
C = 0 + 1E + u
where u is the unobserved error.

i What kinds of factors are contained in u? Are these likely to be correlated with the level
of education of Australian women?
ii Will a simple regression analysis as shown above be sufficient to uncover the effect of
education on fertility, other things remaining the same?
2 Suppose you have two different sample sets of variables x and y and you estimate a simple
linear regression model y = 0 + 1x + u. Your estimated models are yˆ = 1.2 + 0.75x and
yˆ = 1.2 + 0.95x , obtained from the first and second sample, respectively. Draw both the
144 Part 2 » R e g R e s s i o n a n a l y s i s w i t h c R o s s - s e c t i o n a l d a t a
fitted lines in a single graph and compare them.
3 The following table contains the data taken on number of weekly study hours (HOURS)
andwhere,
the final
asmarks (MARKS)
usual in a modelobtained by students
with an intercept, weincan
an assume E(u) =econometrics
introductory unit in a
0. List at least two
university
factorsincontained
South Australia.
in u. Are these likely to have positive or negative correlation with hours?
iii In the equation from part (ii), what should be the sign of 1 if the preparation course is
student MARKS HOURS
effective?
1 58 21
iv In the equation from part (ii), what is the interpretation of 0?
2 69 24
16 In the potential outcomes framework, 3 suppose62 that program26 eligibility is randomly
assigned but participation cannot 4 be enforced. 73 To formally describe
27 this situation, for
each person i, zi is the eligibility 5 indicator and74 xi is the participation
29 indicator. Randomised
eligibility means zi is independent 6 of [yi (0),yi (1)] 62 but xi might not 25 satisfy the independence
assumption. 7 55 25
i Explain why the difference8in means estimator 76 is generally30no longer unbiased.
146 Part 2 » R e g R eii
s s i oIn
n athe
n a lcontext
y s i s w i t hof
c Rao sjob
s - s training
e c t i o n a l program,
data what kind of individual behaviour would cause
i Estimate the relationship between MARKS and HOURS using OLS; that is, obtain the
bias?
intercept and slope estimates in the regression of MARKS on HOURS. Comment on
the direction of the relationship. Does the intercept have a useful interpretation here?
10 Which of the following is correct about R 2 ?
Explain. How much higher is MARKS predicted to be if the HOURS is increased by five Multiple-choice questions provide an
multiple units? choice questions
a R 2 ranges between −1 to +1.
b R 2 is measured as the ratio of sum squared residuals to total sum of squares.
1 ciiConsiderCompute the fitted scatter
the following
R 2 measures
values and
the goodness-of-fit plot residuals
ofoftwo for each
thediscrete
estimated
observation,
variables, and verify that the
model.consumption (C) and income
opportunity to revise and reinforce your
residuals to (approximately) onsum to zero.
(X) (drawn
d
iii
the of
R 2 is measured
What is the
scale).as Based
predicted
one minus
value
thethe
of
information
MARKS when
conveyedsum
ratio of explained
HOURS
by the
=
plot, what
of squares
25?
squares.intercept βˆ10 and estimated slope βˆ1 ? (Note: You need to carefully scrutinise
estimated
to are
totalthe values of
sum
learning.
iv How
the plot andmuch of the
record thevariation
values of intheMARKS for these
dependent eight C
variable students
and theisindependent HOURSX).
explained byvariable ?
11 Assume that you have a sample of 10 observations of x and y, such that:
Explain.
4 ∑are
Suppose you
10
yi =given
5 ∑ xi =10 of consumption
Scatter plot data
the following ∑ xvariables
on two i = 20
x and ∑
2 (C) and income (X)
y: xi yi = 15.
Now9consider
y 2 the following
8 two
3 equations:
11 5 6 1 7 7 6
x 8 2 7 6 2nβˆ10− 2 y6 + 2 0 ∑ i ∑ x9 βˆ i 1 = 0−1 8 8 8
7
2∑ xi2 βˆ1 − 2∑ xi yi + 2∑ xi βˆ0 = 0 .
6
CWhat5 are the values of βˆ0 and βˆ ?
1
a βˆ40 = −0.5, βˆ1 = 1

b βˆ30 = −2, βˆ1 = 1.5
2
c βˆ0 = 0.8, βˆ1 = −1
1
d βˆ0 = 2.5, βˆ1 = −1
0
0 the following
12 Consider 1 2
table 3 random
of two 4 variables
5 x and y6. Suppose
7 you regress
8 y 9
on x. What is the sum of squared residuals ∑ûX i ?
2
( )
a βˆ0 = −2.5 andy βˆ1 = 1.28
1 2 4 3 1 3 4 2
b x βˆ1 =11.24 2
βˆ0 = −1.67 and 3 2 4 1 2 4
ac 0βˆ0 = −4.28 and βˆ1 = .82

bd 9.65
βˆ0 = 1.03 and βˆ1 = −1.28
c 9.772
2 d 1.276 a regression model y = β0 + β1 x i + ui and suppose from a sample of
Consider
10 observations you are provided the following information:
∑y = 234, x = 32, ∑xy = 7138 and ∑x 2 = 11 776 . Put chapter theory into practice with
computer questions
Given this information, what is the predicted value of y, i.e., ŷ, for x = 24?
C1 Economists
a 30.69 have begun to study the patterns in wellbeing data. Blanchflower and
the Computer questions.
b 25.22(2005) examine the relationship between data from the International Social
Oswald
c 36.16
Survey Programme for the year 2002 on a happiness index (happ) and the UN human
d None of the
development above
index (HDI). The happiness index combines information on wellbeing
questions and is on a scale that runs from a low of 1 to a high of 7. The HDI is a welfare
score from 0–1 that combines three indicators: lifespan, educational attainment and
adjusted real income. The data for 31 countries is in the data file HAPP.RAW.
i Estimate the simple regression model
happ = β0 + β1HDI + u
and report your results in equation form along with the number of observations and R2.
ii Find the predicted happ for Australia. How does this compare to the actual value of
happ?
iii Which eight countries appear particularly unhappy? Omit these countries from the
sample and re-estimate the model. Is there a relationship between the HDI and
happiness?

Another random document with
no related content on Scribd:
A. de Lamartine:
Joan of Arc,
1208-1209 (1175-1176).
Lord Mahon:
Historical Essays,
1209 (1177).
J. O’Hagan:
Joan of Arc,
1209 (1177).
T. de Quincey:
Joan of Arc,
1209-1210 (1177-1178).
"Her ways and habits during the year she was in arms are
attested by a multitude of witnesses. Dunois and the Duke of
Alençon bear testimony to what they term her extraordinary
talents for war, and to her perfect fearlessness in action;
but in all other things she was the most simple of creatures.
She wept when she first saw men slain in battle, to think that
they should have died without confession. She wept at the
abominable epithets which the English heaped upon her; but she
was without a trace of vindictiveness. … In her diet she was
abstemious in the extreme, rarely eating until evening, and
then for the most part, of bread and water, sometimes mixed
with wine. In the field, she slept in her armor; but when she
came into a city, she always sought out some honorable matron,
under whose protection she placed herself; and there is
wonderful evidence of the atmosphere of purity which she
diffused around her, her very presence banishing from men’s
hearts all evil thoughts and wishes. Her conversation, when
not of war, was entirely of religion. She confessed often, and
received communion twice in the week."
J. O’HAGAN.
{752}
10. THE EFFECTS OF THE ONE HUNDRED YEARS WAR:
E. E. Crowe:
History of France,
1210 (1178).
H. Hallam:
The Middle Ages,
1211 (1179).
C. W. Oman:
Warwick the King-Maker,
846-847 (819-820).
J. N. Larned:
Europe,
1065-1068 (1037-1040).
11. THE PRAGMATIC SANCTION OF CHARLES VII.

(A. D. 1438):
R. C. Trench:
Church History,
2500 (2440).
M. Creighton:
History of the Papacy,
1210-1211 (1178-1179).
"Such were the chief reforms of its own special grievances

which France wished to establish. It was the first step in the
assertion of the rights of National Churches to arrange for
themselves the details of their own ecclesiastical
organizations."
M. CREIGHTON.
STUDY XVII.
Page references in first 1895 edition in parentheses.
ITALY TO THE END OF THE MIDDLE AGES

(A. D. 1000-1450).
1. GENERAL CONDITIONS AT THE CLOSE OF THE TENTH

CENTURY:
J. Bryce:
The Holy Roman Empire,
1848 (1808).
A. F. Villemain:
Life of Gregory VII.,
2820 (2746).
H. H. Milman:
Latin Christianity,
2820 (2746).
J. C. L. Sismondi:
The Italian Republics,
1848 (1808).
H. Hallam:
The Middle Ages,
1848-1849 (1808-1809).
J. Bryce:
2725 (2652).
L. von Ranke:
History of the World,
2725-2726.
P. Godwin:
History of France,
2078, first column, (2034).
2. THE NORMAN SETTLEMENTS

(A. D. 1000-1100):
A. Thierry:
Conquest of England,
2418 (2366).
Sir F. Palgrave:
History of Normandy, etc.,
2419-2420 (2367-2368).
E. A. Freeman:
The Norman Conquest,
2421-2422 (2369-2370).
Sir F. Palgrave:
History of Normandy,
2422 (2370).
C. Thirlwall:
History of Greece,
2981-2982 (2903-2904).
E. A. Freeman:
Story of Sicily,
2983 (2905).
G. Finlay:
The Byzantine Empire,
2984 (2906).
E. Gibbon:
Decline and Fall,
2984 (2906).
J. Michelet:
History of France,
1849 (1809).
G. Procter:
History of Italy,
1849-1850 (1809-1810).
H. H. Milman:
Latin Christianity,
2821 (2747).
A. H. Johnson:
The Normans in Europe,
1850-1851 (1810-1811).
J. N. Larned:
Europe,
1051 (1023).
3. RISE OF THE FREE CITIES:
P. Godwin:
History of France,
2077-2078 (2033-2034).
J. C. L. Sismondi:
1850 (1810).
Hinschius:
Investiturstreit,
2488-2489 (3794-3796).
H. E. Napier:
Florentine History,
3273 (3157).
(a) Milan.
W. Ihne:
History of Rome,
2746-2747 (2672-2673).
E. Gibbon:
Decline and Fall,
2226 (2182).
G. B. Testa:
War of Frederick I. against Lombardy,
2226 (2182).
T. Hodgkin:
Italy and Her Invaders,
2226-2227 (2182-2183).
(b) Florence.
H. E. Napier:
Florentine History,
1160 (1130).
T. A. Trollope:
Commonwealth of Florence,
1160-1161 (1130-1131).
B. Duffy:
The Tuscan Republics,
1161 (1131).
(c) Pavia.
G. B Niebuhr:
History of Rome,
2070 (2026).
E. Gibbon:
Decline and Fall,
2077 (2033).
(d) Pisa.
L. Pignotti:
History of Tuscany,
2605-2606 (2537-2538).
J. C. L. Sismondi:
Italian Republics,
2606-2607 (2538-2539).
(e) Venice.
E. Gibbon:
Decline and Fall,
3722 (3602).
T. Hodgkin:
Italy and Her Invaders,
3722 (3602).
G. Finlay:
Byzantine Empire,
3722-3723 (3602-3603).
J. C. L. Sismondi:
Italian Republics,
3724-3725 (3604-3605).
4. ESTABLISHMENT OF THE STATES OF THE CHURCH:
J. N. Murphy:
The Chair of Peter,
2492 (2432).
H. E. Napier:
Florentine History,
3273 (3157).
M. Creighton:
History of the Papacy,
2493 (2433).
5. CONDITIONS IN ROME.
J. I. Döllinger:
European History,
2821 (2747).
H. H. Milman:
Latin Christianity,
2821 (2747).
E. Gibbon:
Decline and Fall,
2822 (2748).
6. STRUGGLE OF THE ITALIAN REPUBLICS WITH THE

EMPERORS:
(a) With Frederick I., Barbarossa (A. D. 1154-1183).
O. Browning:
Guelphs and Ghibellines,
1478-1479 (1445-1446).
J. C. L. Sismondi:
1851-1852 (1811-1812).
U. Balzani:
The Popes and the Hohenstaufen,
1852 (1812).
J. C. L. Sismondi:
1852 (1812).
U. Balzani:
1852-1853 (1812-1813).
W. Menzel:
History of Germany,
1853 (1813).
(b) With Frederick the Second

(A. D. 1220-1250).
J. Bryce:
1854 (1814).
J. C. L. Sismondi:
Italian Republics,
1137-1138 (1109-1110).
E. A. Freeman:
European History,
1479 (1446).
E. A. Freeman:
Frederick the Second,
1480, first column (1447).
J. A. Symonds:
The Revival of Learning,
720 (697).
(c) The Results of the Contest.
J. Burckhardt:
The Renaissance in Italy,
1856-1857 (1816-1817).
O. Browning:
Guelfs and Ghibellines,
1856 (1816).
E. Smedley:
History of France,
1858-1859 (1818-1819).
J. A. Symonds:
Florence and the Medici,
1163 (1133).
7. THE GUELFS AND GHIBELLINES:
U. Balzani:
1478 (1445).
H. Hallam:
Middle Ages,
1652 (1614).
Sir A. Halliday:
Annals of House of Hanover,
1652 (1614).
T. A. Trollope:
1857-1858 (1817-1818).
R. W. Church:
Dante,
1858 (1818).
N. Machiavelli:
History of Florence,
1161-1162 (1131-1132).
{753}
O. Browning:
Guelphs and Ghibellines,
1162 (1132).
T. A. Trollope:
The Commonwealth of Florence,
1162-1163 (1132-1133).
8. THE AGE OF THE DESPOTS

(A. D. 1250-1500):
J. Burckhardt:
1856-1857 (1816-1817).
T. A. Trollope:
The Commonwealth of Florence,
1857-1858 (1817-1818).
E. A. Freeman:
Historical Geography of Europe,
1859 (1819).
J. A. Symonds:
1859 (1819).
J. Yeats:
Growth of Commerce,
2249, second column, (2205).
A. von Reumont:
Lorenzo de’ Medici,
2250 (2206).
T. A. Trollope:
2250 (2206).
J. A. Symonds:
The Renaissance,
2463-2464.
J. N. Larned:
Europe,
1074-1045 (1046-1047).
9. CONTINUED CONTESTS BETWEEN THE GUELFS AND

GHIBELLINES:
W. Hunt:
History of Italy,
1860-1861 (1820-1821).
H. E. Napier:
Florentine History,
1861 (1821).
G. Procter:
History of Italy,
1862-1863 (1822-1823).
10. RIENZI; THE LAST OF THE TRIBUNES

(A. D. 1347-1354):
Professor de Vericour:
Rienzi, 2822-2824 (2748-2750).
W. W. Story:
The Castle of St. Angelo,
2824-2825 (2750-2751).
11. THE INFAMOUS "FREE COMPANIES"

(ABOUT A. D. 1340-1390):
T. A. Trollope:
1865-1866 (1825-1826).
W. P. Urquhart:
Life of F. Sforza,
1866 (1826).
Sir John Hawkwood,

1866 (1826).
12. DEVELOPMENT OF THE CITY PRINCIPALITIES:
(a) Florence.
(1) The Passing of the Republic.
J. A. Symonds:
1163 (1133).
C. Balbo:
Life of Dante,
1164 (1134).
W. P. Urquhart:
Life of F. Sforza,
1165 (1135).
T. B. Macaulay:
Machiavelli,
1166 (1136).
G. Boccaccio:
The Decameron,
1166 (1136).
J. E. T. Rogers:
History of Agriculture,
292-293 (283-284).
T. A. Trollope:
1166-1167 (1136-1137).
H. E. Napier:
Florentine History,
1167 (1137).
(2) The Medici.
J. A. Symonds:
1167-1168 (1137-1138).
T. A. Trollope:
1168-1169 (1138-1139).
W. B. Scaife:
Florentine Life,
1169 (1139).
W. Hunt:
History of Italy,
1169 (1139).
A. von Reumont:
1169-1170 (1139-1140).
J. C. L. Sismondi:
Italian Republics,
1170-1171 (1140-1141).
P. Villari:
Machiavelli,
1171-1172 (1141-1142).
Mrs. Oliphant:
Makers of Florence,
1172 (1142).
H. A. Taine:
Italy, Florence, and Venice,
1172-1173 (1142-1143).
(3) Savonarola.
O. T. Hill:
Savonarola’s Triumph of the Cross,
1173-1175 (1143-1145).
H. E. Napier:
Florentine History,
1176 (1146).
J. A. Symonds:
Studies in Italy,
1176-1177 (1146-1147).
Mrs. Oliphant:
Makers of Florence,
1172 (1142).
"Florence was as near a pagan city as it was possible for its

rulers to make it. … Society had never been more dissolute,
more selfish, or more utterly deprived of any higher aim.
Barren scholarship, busy over grammatical questions, and
elegant philosophy, snipping and piecing its logical systems,
formed the top-dressing to that half-brutal,
hall-superstitious ignorance of the poor. The dilettante world
dreamed hazily of a restoration of the worship of the pagan
gods; Cardinal Bembo bade his friend beware of reading St.
Paul’s Epistles, lest their barbarous style should corrupt his
taste. … Thus limited intellectually, the age of Lorenzo was
still more hopeless morally, full of debauchery, cruelty and
corruption, violating oaths, betraying trusts, believing in
nothing but Greek manuscripts, coins, and statues, caring for
nothing but pleasure. This was the world in which Savonarola
found himself."
MRS. OLIPHANT.
(b) Milan.
J. C. L. Sismondi:
Italian Republics,
J. C. L. Sismondi:
J. A. Symonds:
Age of Despots,
2227-2228 (2183-2184).
W. Robertson:
Charles the Fifth,
2228 (2184).
A. von Reumont:
2228-2229 (2184-2185).
(c) Pisa.
J. T. Bent:
Genoa,
2606-2607 (2538-2539).
J. A. Symonds:
Studies in Italy,
50-51 (43-44).
E. Gibbon:
Decline and Fall,
522-523 (508-509).
G. Procter:
History of Italy,
1862-1863 (1822-1823).
W. Hunt:
History of Italy,
1868 (1828).
J. N. Murphy:
The Chair of St. Peter,
2498 (2438).
(d) Genoa.
J. T. Bent:
Genoa,
1452-1453, 2606-2607 (1419-1420, 2538-2539).
J. A. Symonds:
Renaissance in Italy,
J. T. Bent:
Genoa,
1454, 2251-2252 (1421, 2207-2208).
G. B. Malleson:
Genoese History,
1454 (1421).
J. N. Larned:
Venice and Genoa,
3220 (3709).
J. C. L. Sismondi:
Italian Republics,
1454 (1421).
(e) Venice.
G. Finlay:
Byzantine and Greek Empires,
3726 (3606).
The Republic of Venice,

3726 (3606).
E. Pears:
The Fall of Constantinople,
3726 (3606).
W. C. Hazlitt:
The Venetian Republic,
3727 (3607).
J. Yeats:
The Growth of Commerce,
3727 (3607).
G. Finlay:
Byzantine and Greek Empires,
523-524 (509-510).
F. A. Parker:
Fleets of the World,
3728 (3608).
J. T. Bent:
Genoa,
3729 (3609).
J. C. L. Sismondi:
Italian Republics,
1869 (1829).
STUDY XVIII.
Page references in first 1895 edition in parentheses.
THE EARLY CHRISTIAN CHURCH: FROM PENTECOST TO

GREGORY THE GREAT
(A. D. 30(?)-600).
1. JUDÆA AT THE BEGINNING OF THE CHRISTIAN ERA:
E. de Pressensé:
Jesus Christ,
1961-1962 (1920-1921).
E. Schürer:
The Jewish People,
1678 (1639).
A. Edersheim:
Life of Jesus,
446 (432).
H. W. Hulbert:
Historical Geography,
446 (432).
{754}
2. HEROD AND THE HERODIANS

(B. C. 40-A. D. 44):
T. Keim:
Jesus of Nazara,
1958-1959 (1917-1918).
T. Mommsen:
History of Rome,
1960 (1919).
H. H. Milman:
History of the Jews,
1960 (1919).
3. THE BIRTH OF JESUS:

T. Keim:
Jesus of Nazara,
1960-1961 (1919-1920).
W. Hales:
Analysis of Chronology,
1011 (984).
4. PENTECOST, AND THE ESTABLISHMENT OF THE FIRST

CHURCHES:
G. Y. Lechler:
The Apostolic Times,
447 (433).
A. Sabatier:
The Apostle Paul,
447 (433).
J. B. Lightfoot:
The Apostolic Age,
448 (434).
W. Moeller:
The Christian Church,
448 (434).
J. E. Wiltsch:
Statistics of the Church,
448 (434).
J. B. Lightfoot:
The Apostolic Age,
449 (435).
W. Moeller:
449 (435).
5. THE APOSTOLIC PERIOD (A. D. 30(?)-100):
(a) The Church at Antioch.
C. Thirlwall:
History of Greece,
2107, 2960 (2063, 2883).
W. Moeller:
448 (434).
J. J. von Döllinger:
European History,
449 (435).
W. M. Ramsay:
The Church in the Roman Empire,
449 (435).
W. Moeller:
449 (435).
B. Weiss:
Introduction to the New Testament,
450 (436).
G. B. Brown:
From Schola to Cathedral,
450 (436).
(b) The Missions of St. Paul.

G. P. Fisher:
450 (436).
A. Sabatier;
The Apostle Paul,
450-451 (436-437).
A. Sabatier;
The Apostle Paul,
J. B. Lightfoot:
Biblical Essays,
451 (437).
W. M. Ramsay:
451 (437).
C. T. Cruttwell:
Literary History of Early Christianity,
191-192 (184-185).
(c) The Church at Rome.
W. Moeller:
453 (439).
G. Salmon:
Infallibility of the Church,
2476 (2417).
J. J. I. Döllinger:
History of the Church,
2476-2477 (2417-2418).
F. W. Farrar:
Early Days of Christianity,
2781-2782 (2707-2708).
J. B. Lightfoot:
The Apostolic Age,
453 (439).
(d) The Church at Alexandria.
R. S. Poole:
The Cities of Egypt,
44 (37).
E. Kirkpatrick:
Development of Superior Education,
708 (685).
A. Neander:
History of the Christian Church,
452 (438).
J. P. Mahaffy:
Alexander’s Empire,
2973 (2896).
(e) The Destruction of Jerusalem.
J. B. Lightfoot:
The Apostolic Age,
449 (435).
C. Merivale:
History of the Romans,
1962 (1921).
Besant and Palmer:
Jerusalem,
1963 (1922).
H. H. Milman:
History of the Jews,
1963 (1922).
J. B. Lightfoot:
The Apostolic Age,
461 (447).
(f) St. John, and the Church at Ephesus.
E. Abbott:
History of Greece,
J. T. Wood:
Discoveries at Ephesus,
1008-1009 (981-2)
C. Merivale:
History of the Romans,
1009 (982).
J. B. Lightfoot;
Biblical Essays,
451-452 (437-438).
"For Christians are not distinguished from the rest of mankind

either in locality or in speech, or in customs. … They dwell
in their own countries as the lot of each is cast, but only as
sojourners; they bear their share in all things as citizens,
and they endure all hardships as strangers. Every foreign
country is a fatherland to him and every fatherland is
foreign. … Their existence is on earth, but their citizenship
is in heaven. They obey the established laws, and they surpass
the laws in their own lives. They love all men, and they are
persecuted by all. War is urged against them as aliens by the
Jews, and persecution is carried on against them by the
Greeks, and yet those that hate them cannot tell the reason of
their hostility."
THE EPISTLE TO DIOGNETUS (ABOUT A. D. 150).
6. THE PERIOD OF CHURCH DEVELOPMENT (A. D. 100-312):
G. B. Brown:
455 (441).
B. F. Westcott:
Religious Thought in the West,
453-454, 454 (439-440, 440).
J. F. Hurst:
454 (440).
G. Uhlhorn:
Conflict of Christianity with Heathenism,
454 (440).
W. M. Ramsay:
455 (441).
J. H. Kurtz:
Church History,
457 (443).
G. A. Jackson:
The Fathers of the Third Century,
457 (443).
See Map between pages 446-7 (432-3),
and Appendix D, 3806-3810 (End of Volume I.).
7. CHARACTERISTICS OF EARLY CHURCH AND CHRISTIANS:
G. Uhlhorn:
Conflict of Christianity with Heathenism,
454 (440).
J. B. Lightfoot:
Translation Epistle to Diognetus,
454 (440).
R. W. Church:
Gifts of Civilization,
455 (441).
J. B. Lightfoot:
Apostolic Age,
457 (443).
G. P. Fisher:
Christian Church,
459 (445).
W. M. Ramsay:
456 (442).
H. Hayman:
Diocesan Synods,
456 (442).
W. Moeller:
457 (443) .
G. A. Jackson:
Fathers of the Third Century,
457 (443).
J. H. Kurtz:
Church History,
459 (445).
8. THE RISE OF ECCLESIASTICISM:
W. D. Killen:
The Old Catholic Church,
458 (444).
J. B. Lightfoot:
The Apostolic Age,
458 (444).
C. Gore:
The Mission of the Church,
458 (444).
A. Neander:
The Christian Religion,
458 (444).
9. GROWTH OF GREAT CHURCH CENTRES:
F. W. Puller:
Primitive Saints,
458 (444).
(a) Alexandria.
C. T. Cruttwell:
459-460 (445-446).
J. B. Heard:
Alexandrian Theology,
460 (446).
W. Moeller:
Christian Church,
460 (446).
C. Bigg:
The Christian Platonists,
460-461 (446-447).
F. C. Baur:
Church of the First Three Centuries,
1589 (1551).
(b) Rome.
W. Moeller:
462 (448).
R. Lanciani:
Pagan and Christian Rome,
462-463 (448-449).
E. de Pressensé:
Early Years of Christianity,
463 (449).
(c) Carthage.
C. T. Cruttwell:
461-462 (447-448).
J. I. von Döllinger:
European History,
462 (448).
{755}
10. THE PERSECUTIONS:
G. Uhlhorn:
The Conflict of Christianity with Heathenism,
456 (442).
G. B. Brown:
455 (442).
(а) Under Nero

(A. D. 64-68).
F. W. Farrar:
Early Days of Christianity,
2781-2782 (2707-2708).
(b) Under Domitian

(A. D. 93-96).
V. Duruy:
History of Rome,
2784 (2710).
(c) Under Trajan

(A. D. 112-116).
R. W. Browne:
History of Rome,
2786, first column, (2712).

(FREE PDF Sample) (Ebook PDF) Introductory Econometrics: Asia-Pacific 2nd Edition Ebooks

Uploaded by

Copyright:

Available Formats

(FREE PDF Sample) (Ebook PDF) Introductory Econometrics: Asia-Pacific 2nd Edition Ebooks

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

(FREE PDF Sample) (Ebook PDF) Introductory Econometrics: Asia-Pacific 2nd Edition Ebooks

Uploaded by

Copyright:

Available Formats

Full download ebook at ebooksecure.

(eBook PDF) Introductory Econometrics: Asia-

Download more ebook from https://ebooksecure.com

(eBook PDF) Introductory Econometrics Asia Pacific

(eBook PDF) ACCT2 Managerial 2nd Asia Pacific Edition

(eBook PDF) Managerial Accounting Asia-Pacific Edition

(eBook PDF) ACCT2 Financial 2nd Asia-Pacific Edition by

(eBook PDF) Marketing Principles 2nd Asia Pacific

(eBook PDF) Principles of Anatomy and Physiology 2nd

(eBook PDF) Managerial Accounting 1st Asia Pacific

(Original PDF) Marketing Research 3rd Asia-Pacific

Chapter 03 fundamentals of statIstIcs: a revIew 46

0 Wooldridge1e_50839_txt_3pp.indd 7 16/09/20 10:22 AM

3.8 Confidence intervals and hypothesis testing 79

0 Wooldridge1e_50839_txt_3pp.indd 8 16/09/20 10:22 AM

Chapter 05 multIple regressIon analysIs: estImatIon 150

Chapter 06 multIple regressIon analysIs: Inference 195

0 Wooldridge1e_50839_txt_3pp.indd 9 16/09/20 10:22 AM

» Testing other hypotheses about βj 205

Chapter 07 model specIfIcatIon 244

0 Wooldridge1e_50839_txt_3pp.indd 10 16/09/20 10:22 AM

Chapter 08 multIple regressIon analysIs wIth qualItatIve

Chapter 09 heteroscedastIcIty 319

0 Wooldridge1e_50839_txt_3pp.indd 11 16/09/20 10:22 AM

Part regressIon analysIs wIth tIme serIes data 349

Chapter 11 serIal correlatIon and heteroscedastIcIty

0 Wooldridge1e_50839_txt_3pp.indd 12 16/09/20 10:22 AM

Chapter 12 modellIng hIghly persIstent tIme serIes data 424

Part addItIonal topIcs 445

0 Wooldridge1e_50839_txt_3pp.indd 13 16/09/20 10:22 AM

13.3 Fixed effects estimator 449

Chapter 14 lImIted dependent varIaBle models

0 Wooldridge1e_50839_txt_3pp.indd 14 16/09/20 10:22 AM

Chapter 15 an IntroductIon to machIne learnIng

APPENDIX A Statistical tables 539

0 Wooldridge1e_50839_txt_3pp.indd 15 16/09/20 10:22 AM

0 Wooldridge1e_50839_txt_3pp.indd 16 16/09/20 10:22 AM

majority of econometrics texts introduce a complete The AsiA–PAcific ediTion

0 Wooldridge1e_50839_txt_3pp.indd 17 24/09/20 5:01 PM

Part openers introduce each

the nature of econometrIcs Identify the key concepts you

The Chapter overview

1.1 what Is econometrIcs?

examPle Job training and worker productivity

gaming machines that demonstrates summAry

0 Wooldridge1e_50839_txt_3pp.indd 19 Key terms 16/09/20 10:22 AM

Test your knowledge and consolidate

where u is the unobserved error.

a βˆ40 = −0.5, βˆ1 = 1

ac 0βˆ0 = −4.28 and βˆ1 = .82

0 Wooldridge1e_50839_txt_3pp.indd 20 16/09/20 10:22 AM

10. THE EFFECTS OF THE ONE HUNDRED YEARS WAR:

11. THE PRAGMATIC SANCTION OF CHARLES VII.

"Such were the chief reforms of its own special grievances

ITALY TO THE END OF THE MIDDLE AGES

1. GENERAL CONDITIONS AT THE CLOSE OF THE TENTH

2. THE NORMAN SETTLEMENTS

3. RISE OF THE FREE CITIES:

4. ESTABLISHMENT OF THE STATES OF THE CHURCH: