k

Principles of
Econometrics
k Fifth Edition k

R. CARTER HILL
Louisiana State University

WILLIAM E. GRIFFITHS
University of Melbourne

GUAY C. LIM
University of Melbourne

k
 k

Brief Contents
PREFACE v 10 Endogenous Regressors and
LIST OF EXAMPLES xxi Moment-Based Estimation 481
1 An Introduction to Econometrics 1 11 Simultaneous Equations Models 531
Probability Primer 15
12 Regression with Time-Series Data:
2 The Simple Linear Regression Nonstationary Variables 563
Model 46
13 Vector Error Correction and Vector
3 Interval Estimation and Hypothesis Autoregressive Models 597
Testing 112
14 Time-Varying Volatility and ARCH
4 Prediction, Goodness-of-Fit, and Models 614
Modeling Issues 152
15 Panel Data Models 634
5 The Multiple Regression Model 196
16 Qualitative and Limited Dependent
6 Further Inference in the Multiple Variable Models 681
k Regression Model 260 k
APPENDIX A Mathematical Tools 748

7 Using Indicator Variables 317 APPENDIXB Probability Concepts 768

8 Heteroskedasticity 368 APPENDIXC Review of Statistical Inference 812

APPENDIXD Statistical Tables 862

9 Regression with Time-Series Data:
Stationary Variables 417 INDEX 869

iv

k
 k

Table of Contents
PREFACE v P.5.6 Covariance Between Two Random
LIST OF EXAMPLES xxi Variables 27
P.6 Conditioning 29
1 An Introduction to P.6.1 Conditional Expectation 30
P.6.2 Conditional Variance 31
Econometrics 1 P.6.3 Iterated Expectations 32
P.6.4 Variance Decomposition 33
1.1 Why Study Econometrics? 1 P.6.5 Covariance Decomposition 34
1.2 What Is Econometrics About? 2 P.7 The Normal Distribution 34
1.2.1 Some Examples 3 P.7.1 The Bivariate Normal Distribution 37
1.3 The Econometric Model 4 P.8 Exercises 39
1.3.1 Causality and Prediction 5
1.4 How Are Data Generated?
1.4.1 Experimental Data 6
5 2 The Simple Linear Regression
1.4.2 Quasi-Experimental Data 6 Model 46
1.4.3 Nonexperimental Data 7
1.5 Economic Data Types 7 2.1 An Economic Model 47
1.5.1 Time-Series Data 7 2.2 An Econometric Model 49
1.5.2 Cross-Section Data 8 2.2.1 Data Generating Process 51
1.5.3 Panel or Longitudinal Data 9 2.2.2 The Random Error and Strict
k Exogeneity 52
k
1.6 The Research Process 9
2.2.3 The Regression Function 53
1.7 Writing an Empirical Research Paper 11 2.2.4 Random Error Variation 54
1.7.1 Writing a Research Proposal 11 2.2.5 Variation in x 56
1.7.2 A Format for Writing a Research Report 11 2.2.6 Error Normality 56
1.8 Sources of Economic Data 13 2.2.7 Generalizing the Exogeneity
1.8.1 Links to Economic Data on the Internet 13 Assumption 56
1.8.2 Interpreting Economic Data 14 2.2.8 Error Correlation 57
1.8.3 Obtaining the Data 14 2.2.9 Summarizing the Assumptions 58
2.3 Estimating the Regression Parameters 59
Probability Primer 15
2.3.1 The Least Squares Principle 61
P.1 Random Variables 16 2.3.2 Other Economic Models 65
P.2 Probability Distributions 17 2.4 Assessing the Least Squares Estimators 66
P.3 Joint, Marginal, and Conditional 2.4.1 The Estimator b2 67
Probabilities 20 2.4.2 The Expected Values of b1 and b2 68
P.3.1 Marginal Distributions 20 2.4.3 Sampling Variation 69
P.3.2 Conditional Probability 21 2.4.4 The Variances and Covariance of b1
P.3.3 Statistical Independence 21 and b2 69
P.4 A Digression: Summation Notation 22 2.5 The Gauss–Markov Theorem 72
P.5 Properties of Probability Distributions 23 2.6 The Probability Distributions of the Least
P.5.1 Expected Value of a Random Variable 24 Squares Estimators 73
P.5.2 Conditional Expectation 25
2.7 Estimating the Variance of the Error Term 74
P.5.3 Rules for Expected Values 25
2.7.1 Estimating the Variances and Covariance of
P.5.4 Variance of a Random Variable 26
the Least Squares Estimators 74
P.5.5 Expected Values of Several Random
2.7.2 Interpreting the Standard Errors 76
Variables 27
xii

k
 k

Table of Contents xiii

2.8 Estimating Nonlinear Relationships 77 3.2 Hypothesis Tests 118

2.8.1 Quadratic Functions 77 3.2.1 The Null Hypothesis 118
2.8.2 Using a Quadratic Model 77 3.2.2 The Alternative Hypothesis 118
2.8.3 A Log-Linear Function 79 3.2.3 The Test Statistic 119
2.8.4 Using a Log-Linear Model 80 3.2.4 The Rejection Region 119
2.8.5 Choosing a Functional Form 82 3.2.5 A Conclusion 120
2.9 Regression with Indicator Variables 82 3.3 Rejection Regions for Specific
2.10 The Independent Variable 84 Alternatives 120
2.10.1 Random and Independent x 84 3.3.1 One-Tail Tests with Alternative
2.10.2 Random and Strictly Exogenous x 86 ‘‘Greater Than’’ (>) 120
2.10.3 Random Sampling 87 3.3.2 One-Tail Tests with Alternative
‘‘Less Than’’ (<) 121
2.11 Exercises 89 3.3.3 Two-Tail Tests with Alternative
2.11.1 Problems 89
‘‘Not Equal To’’ (≠) 122
2.11.2 Computer Exercises 93
3.4 Examples of Hypothesis Tests 123
Appendix 2A Derivation of the Least Squares
Estimates 98 3.5 The p-Value 126
Appendix 2B Deviation from the Mean Form 3.6 Linear Combinations of Parameters 129
of b2 99 3.6.1 Testing a Linear Combination of
Parameters 131
Appendix 2C b2 Is a Linear Estimator 100
3.7 Exercises 133
Appendix 2D Derivation of Theoretical Expression 3.7.1 Problems 133
for b2 100 3.7.2 Computer Exercises 139
Appendix 2E Deriving the Conditional Variance Appendix 3A Derivation of the t-Distribution 144
k of b2 100 k
Appendix 3B Distribution of the t-Statistic under
Appendix 2F Proof of the Gauss–Markov H1 145
Theorem 102
Appendix 3C Monte Carlo Simulation 147
Appendix 2G Proofs of Results Introduced in 3C.1 Sampling Properties of Interval
Section 2.10 103 Estimators 148
2G.1 The Implications of Strict Exogeneity 103 3C.2 Sampling Properties of Hypothesis
2G.2 The Random and Independent x Case 103 Tests 149
2G.3 The Random and Strictly Exogenous x 3C.3 Choosing the Number of Monte Carlo
Case 105 Samples 149
2G.4 Random Sampling 106 3C.4 Random-x Monte Carlo Results 150
Appendix 2H Monte Carlo Simulation 106
2H.1
2H.2
The Regression Function 106
The Random Error 107
4 Prediction, Goodness-of-Fit,
2H.3 Theoretically True Values 107 and Modeling Issues 152
2H.4 Creating a Sample of Data 108
2H.5 Monte Carlo Objectives 109 4.1 Least Squares Prediction 153
2H.6 Monte Carlo Results 109 4.2 Measuring Goodness-of-Fit 156
2H.7 Random-x Monte Carlo Results 110 4.2.1 Correlation Analysis 158
4.2.2 Correlation Analysis and R2 158
3 Interval Estimation and 4.3 Modeling Issues 160
Hypothesis Testing 112 4.3.1 The Effects of Scaling the Data 160
4.3.2 Choosing a Functional Form 161
4.3.3 A Linear-Log Food Expenditure Model 163
3.1 Interval Estimation 113
4.3.4 Using Diagnostic Residual Plots 165
3.1.1 The t-Distribution 113
4.3.5 Are the Regression Errors Normally
3.1.2 Obtaining Interval Estimates 115
Distributed? 167
3.1.3 The Sampling Context 116

k
 k

xiv Table of Contents

4.3.6 Identifying Influential Observations 169 5.5.2 One-Tail Hypothesis Testing for a Single
4.4 Polynomial Models 171 Coefficient 220
4.4.1 Quadratic and Cubic Equations 171 5.5.3 Hypothesis Testing for a Linear Combination
of Coefficients 221
4.5 Log-Linear Models 173
4.5.1 Prediction in the Log-Linear Model 175 5.6 Nonlinear Relationships 222
4.5.2 A Generalized R2 Measure 176 5.7 Large Sample Properties of the Least Squares
4.5.3 Prediction Intervals in the Log-Linear Estimator 227
Model 177 5.7.1 Consistency 227
4.6 Log-Log Models 177 5.7.2 Asymptotic Normality 229
5.7.3 Relaxing Assumptions 230
4.7 Exercises 179 5.7.4 Inference for a Nonlinear Function of
4.7.1 Problems 179
Coefficients 232
4.7.2 Computer Exercises 185
5.8 Exercises 234
Appendix 4A Development of a Prediction 5.8.1 Problems 234
Interval 192 5.8.2 Computer Exercises 240
Appendix 4B The Sum of Squares Appendix 5A Derivation of Least Squares
Decomposition 193 Estimators 247
Appendix 4C Mean Squared Error: Estimation and
Appendix 5B The Delta Method 248
Prediction 193 5B.1 Nonlinear Function of a Single
Parameter 248
5 The Multiple Regression 5B.2 Nonlinear Function of Two Parameters 249
Model 196 Appendix 5C Monte Carlo Simulation 250
5C.1 Least Squares Estimation with Chi-Square
k 5.1 Introduction 197 Errors 250 k
5.1.1 The Economic Model 197 5C.2 Monte Carlo Simulation of the Delta
5.1.2 The Econometric Model 198 Method 252
5.1.3 The General Model 202 Appendix 5D Bootstrapping 254
5.1.4 Assumptions of the Multiple Regression 5D.1 Resampling 255
Model 203 5D.2 Bootstrap Bias Estimate 256
5.2 Estimating the Parameters of the Multiple 5D.3 Bootstrap Standard Error 256
Regression Model 205 5D.4 Bootstrap Percentile Interval Estimate 257
5.2.1 Least Squares Estimation Procedure 205 5D.5 Asymptotic Refinement 258
5.2.2 Estimating the Error Variance σ2 207
5.2.3 Measuring Goodness-of-Fit 208 6 Further Inference in the Multiple
5.2.4 Frisch–Waugh–Lovell (FWL) Theorem 209
Regression Model 260
5.3 Finite Sample Properties of the Least Squares
Estimator 211 6.1 Testing Joint Hypotheses: The F-test 261
5.3.1 The Variances and Covariances of the Least 6.1.1 Testing the Significance of the Model 264
Squares Estimators 212 6.1.2 The Relationship Between t- and
5.3.2 The Distribution of the Least Squares F-Tests 265
Estimators 214 6.1.3 More General F-Tests 267
5.4 Interval Estimation 216 6.1.4 Using Computer Software 268
5.4.1 Interval Estimation for a Single 6.1.5 Large Sample Tests 269
Coefficient 216 6.2 The Use of Nonsample Information 271
5.4.2 Interval Estimation for a Linear Combination
6.3 Model Specification 273
of Coefficients 217
6.3.1 Causality versus Prediction 273
5.5 Hypothesis Testing 218 6.3.2 Omitted Variables 275
5.5.1 Testing the Significance of a Single 6.3.3 Irrelevant Variables 277
Coefficient 219

k
 k

Table of Contents xv

6.3.4 Control Variables 278 7.7 Exercises 351

6.3.5 Choosing a Model 280 7.7.1 Problems 351
6.3.6 RESET 281 7.7.2 Computer Exercises 358
6.4 Prediction 282 Appendix 7A Details of Log-Linear Model
6.4.1 Predictive Model Selection Interpretation 366
Criteria 285
Appendix 7B Derivation of the
6.5 Poor Data, Collinearity, and Differences-in-Differences
Insignificance 288 Estimator 366
6.5.1 The Consequences of Collinearity 289
Appendix 7C The Overlap Assumption:
6.5.2 Identifying and Mitigating Collinearity 290
Details 367
6.5.3 Investigating Influential Observations 293
6.6 Nonlinear Least Squares 294
8 Heteroskedasticity 368
6.7 Exercises 297
6.7.1 Problems 297 8.1 The Nature of Heteroskedasticity 369
6.7.2 Computer Exercises 303
8.2 Heteroskedasticity in the Multiple Regression
Appendix 6A The Statistical Power of F-Tests 311
Model 370
Appendix 6B Further Results from the FWL 8.2.1 The Heteroskedastic Regression Model 371
Theorem 315 8.2.2 Heteroskedasticity Consequences for the OLS
Estimator 373
7 Using Indicator Variables 317 8.3 Heteroskedasticity Robust Variance
Estimator 374
7.1 Indicator Variables 318 8.4 Generalized Least Squares: Known Form
7.1.1 Intercept Indicator Variables 318 of Variance 375
k k
7.1.2 Slope-Indicator Variables 320 8.4.1 Transforming the Model: Proportional
7.2 Applying Indicator Variables 323 Heteroskedasticity 375
7.2.1 Interactions Between Qualitative 8.4.2 Weighted Least Squares: Proportional
Factors 323 Heteroskedasticity 377
7.2.2 Qualitative Factors with Several 8.5 Generalized Least Squares: Unknown Form
Categories 324
of Variance 379
7.2.3 Testing the Equivalence of Two
8.5.1 Estimating the Multiplicative Model 381
Regressions 326
7.2.4 Controlling for Time 328 8.6 Detecting Heteroskedasticity 383
8.6.1 Residual Plots 384
7.3 Log-Linear Models 329 8.6.2 The Goldfeld–Quandt Test 384
7.3.1 A Rough Calculation 330
8.6.3 A General Test for Conditional
7.3.2 An Exact Calculation 330
Heteroskedasticity 385
7.4 The Linear Probability Model 331 8.6.4 The White Test 387
7.5 Treatment Effects 332 8.6.5 Model Specification and
7.5.1 The Difference Estimator 334 Heteroskedasticity 388
7.5.2 Analysis of the Difference Estimator 334 8.7 Heteroskedasticity in the Linear Probability
7.5.3 The Differences-in-Differences Model 390
Estimator 338
8.8 Exercises 391
7.6 Treatment Effects and Causal Modeling 342 8.8.1 Problems 391
7.6.1 The Nature of Causal Effects 342 8.8.2 Computer Exercises 401
7.6.2 Treatment Effect Models 343
Appendix 8A Properties of the Least Squares
7.6.3 Decomposing the Treatment Effect 344
Estimator 407
7.6.4 Introducing Control Variables 345
7.6.5 The Overlap Assumption 347 Appendix 8B Lagrange Multiplier Tests for
7.6.6 Regression Discontinuity Designs 347 Heteroskedasticity 408

k
 k

xvi Table of Contents

Appendix 8C Properties of the Least Squares 10.1.2 Why Least Squares Estimation
Residuals 410 Fails 484
8C.1 Details of Multiplicative Heteroskedasticity 10.1.3 Proving the Inconsistency
Model 411 of OLS 486
Appendix 8D Alternative Robust Sandwich 10.2 Cases in Which x and e are Contemporaneously
Estimators 411 Correlated 487
10.2.1 Measurement Error 487
Appendix 8E Monte Carlo Evidence: OLS, GLS, and
10.2.2 Simultaneous Equations Bias 488
FGLS 414
10.2.3 Lagged-Dependent Variable Models with
Serial Correlation 489
9 Regression with Time-Series 10.2.4 Omitted Variables 489
Data: Stationary Variables 417 10.3 Estimators Based on the Method
of Moments 490
9.1 Introduction 418 10.3.1 Method of Moments Estimation of a
9.1.1 Modeling Dynamic Relationships 420 Population Mean and Variance 490
9.1.2 Autocorrelations 424 10.3.2 Method of Moments Estimation in the Simple
Regression Model 491
9.2 Stationarity and Weak Dependence 427
10.3.3 Instrumental Variables Estimation in the
9.3 Forecasting 430 Simple Regression Model 492
9.3.1 Forecast Intervals and Standard Errors 433 10.3.4 The Importance of Using Strong
9.3.2 Assumptions for Forecasting 435 Instruments 493
9.3.3 Selecting Lag Lengths 436 10.3.5 Proving the Consistency of the IV
9.3.4 Testing for Granger Causality 437 Estimator 494
9.4 Testing for Serially Correlated Errors 438 10.3.6 IV Estimation Using Two-Stage Least Squares
9.4.1 Checking the Correlogram of the Least (2SLS) 495
k k
Squares Residuals 439 10.3.7 Using Surplus Moment Conditions 496
9.4.2 Lagrange Multiplier Test 440 10.3.8 Instrumental Variables Estimation in the
9.4.3 Durbin–Watson Test 443 Multiple Regression Model 498
10.3.9 Assessing Instrument Strength Using the
9.5 Time-Series Regressions for Policy
First-Stage Model 500
Analysis 443
10.3.10 Instrumental Variables Estimation in a
9.5.1 Finite Distributed Lags 445
General Model 502
9.5.2 HAC Standard Errors 448
10.3.11 Additional Issues When Using IV
9.5.3 Estimation with AR(1) Errors 452
Estimation 504
9.5.4 Infinite Distributed Lags 456
10.4 Specification Tests 505
9.6 Exercises 463
10.4.1 The Hausman Test for Endogeneity 505
9.6.1 Problems 463
10.4.2 The Logic of the Hausman Test 507
9.6.2 Computer Exercises 468
10.4.3 Testing Instrument Validity 508
Appendix 9A The Durbin–Watson Test 476 10.5 Exercises 510
9A.1 The Durbin–Watson Bounds Test 478
10.5.1 Problems 510
Appendix 9B Properties of an AR(1) Error 479 10.5.2 Computer Exercises 516
Appendix 10A Testing for Weak Instruments 520
10 Endogenous Regressors and 10A.1 A Test for Weak Identification 521
10A.2 Testing for Weak Identification:
Moment-Based Estimation 481
Conclusions 525

10.1 Least Squares Estimation with Endogenous Appendix 10B Monte Carlo Simulation 525
Regressors 482 10B.1 Illustrations Using Simulated Data 526
10.1.1 Large Sample Properties of the OLS 10B.2 The Sampling Properties of
Estimator 483 IV/2SLS 528

k
 k

Table of Contents xvii

11 Simultaneous Equations 12.7 Exercises 588

12.7.1 Problems 588
Models 531 12.7.2 Computer Exercises 592

11.1 A Supply and Demand Model 532

13 Vector Error Correction and
11.2 The Reduced-Form Equations 534
Vector Autoregressive
11.3 The Failure of Least Squares Estimation 535
11.3.1 Proving the Failure of OLS 535 Models 597
11.4 The Identification Problem 536
13.1 VEC and VAR Models 598
11.5 Two-Stage Least Squares Estimation 538
13.2 Estimating a Vector Error Correction
11.5.1 The General Two-Stage Least Squares
Model 600
Estimation Procedure 539
11.5.2 The Properties of the Two-Stage Least 13.3 Estimating a VAR Model 601
Squares Estimator 540 13.4 Impulse Responses and Variance
11.6 Exercises 545 Decompositions 603
11.6.1 Problems 545 13.4.1 Impulse Response Functions 603
11.6.2 Computer Exercises 551 13.4.2 Forecast Error Variance
Appendix 11A 2SLS Alternatives 557 Decompositions 605
11A.1 The k-Class of Estimators 557 13.5 Exercises 607
11A.2 The LIML Estimator 558 13.5.1 Problems 607
11A.3 Monte Carlo Simulation Results 562 13.5.2 Computer Exercises 608
Appendix 13A The Identification Problem 612
12 Regression with Time-Series
k k
Data: Nonstationary 14 Time-Varying Volatility and
Variables 563 ARCH Models 614

12.1 Stationary and Nonstationary Variables 564 14.1 The ARCH Model 615
12.1.1 Trend Stationary Variables 567 14.2 Time-Varying Volatility 616
12.1.2 The First-Order Autoregressive Model 570
12.1.3 Random Walk Models 572
14.3 Testing, Estimating, and Forecasting 620

12.2 Consequences of Stochastic Trends 574 14.4 Extensions 622

14.4.1 The GARCH Model—Generalized
12.3 Unit Root Tests for Stationarity 576 ARCH 622
12.3.1 Unit Roots 576 14.4.2 Allowing for an Asymmetric Effect 623
12.3.2 Dickey–Fuller Tests 577 14.4.3 GARCH-in-Mean and Time-Varying Risk
12.3.3 Dickey–Fuller Test with Intercept and No Premium 624
Trend 577 14.4.4 Other Developments 625
12.3.4 Dickey–Fuller Test with Intercept and
Trend 579
14.5 Exercises 626
14.5.1 Problems 626
12.3.5 Dickey–Fuller Test with No Intercept and No
14.5.2 Computer Exercises 627
Trend 580
12.3.6 Order of Integration 581
12.3.7 Other Unit Root Tests 582 15 Panel Data Models 634
12.4 Cointegration 582
12.4.1 The Error Correction Model 584 15.1 The Panel Data Regression Function 636
15.1.1 Further Discussion of Unobserved
12.5 Regression When There Is No
Heterogeneity 638
Cointegration 585 15.1.2 The Panel Data Regression Exogeneity
12.6 Summary 587 Assumption 639

k
 k

xviii Table of Contents

15.1.3 Using OLS to Estimate the Panel Data 16.2.9 Binary Choice Models with a Binary
Regression 639 Endogenous Variable 699
15.2 The Fixed Effects Estimator 640 16.2.10 Binary Endogenous Explanatory
15.2.1 The Difference Estimator: T = 2 640 Variables 700
15.2.2 The Within Estimator: T = 2 642 16.2.11 Binary Choice Models and Panel
15.2.3 The Within Estimator: T > 2 643 Data 701
15.2.4 The Least Squares Dummy Variable 16.3 Multinomial Logit 702
Model 644 16.3.1 Multinomial Logit Choice
15.3 Panel Data Regression Error Assumptions 646 Probabilities 703
15.3.1 OLS Estimation with Cluster-Robust 16.3.2 Maximum Likelihood Estimation 703
Standard Errors 648 16.3.3 Multinomial Logit Postestimation
15.3.2 Fixed Effects Estimation with Cluster-Robust Analysis 704
Standard Errors 650 16.4 Conditional Logit 707
15.4 The Random Effects Estimator 651 16.4.1 Conditional Logit Choice
15.4.1 Testing for Random Effects 653 Probabilities 707
15.4.2 A Hausman Test for Endogeneity in the 16.4.2 Conditional Logit Postestimation
Random Effects Model 654 Analysis 708
15.4.3 A Regression-Based Hausman Test 656 16.5 Ordered Choice Models 709
15.4.4 The Hausman–Taylor Estimator 658 16.5.1 Ordinal Probit Choice Probabilities 710
15.4.5 Summarizing Panel Data Assumptions 660 16.5.2 Ordered Probit Estimation and
15.4.6 Summarizing and Extending Panel Data Interpretation 711
Model Estimation 661 16.6 Models for Count Data 713
15.5 Exercises 663 16.6.1 Maximum Likelihood Estimation of the
15.5.1 Problems 663 Poisson Regression Model 713
k 15.5.2 Computer Exercises 670 16.6.2 Interpreting the Poisson Regression k
Appendix 15A Cluster-Robust Standard Errors: Model 714
Some Details 677 16.7 Limited Dependent Variables 717
16.7.1 Maximum Likelihood Estimation
Appendix 15B Estimation of Error
of the Simple Linear Regression
Components 679
Model 717
16.7.2 Truncated Regression 718
16 Qualitative and Limited 16.7.3 Censored Samples and Regression 718
Dependent Variable Models 681 16.7.4 Tobit Model Interpretation 720
16.7.5 Sample Selection 723
16.1 Introducing Models with Binary Dependent 16.8 Exercises 725
Variables 682 16.8.1 Problems 725
16.1.1 The Linear Probability Model 683 16.8.2 Computer Exercises 733

16.2 Modeling Binary Choices 685 Appendix 16A Probit Marginal Effects: Details 739
16.2.1 The Probit Model for Binary 16A.1 Standard Error of Marginal Effect at a Given
Choice 686 Point 739
16.2.2 Interpreting the Probit Model 687 16A.2 Standard Error of Average Marginal
16.2.3 Maximum Likelihood Estimation of the Probit Effect 740
Model 690 Appendix 16B Random Utility Models 741
16.2.4 The Logit Model for Binary Choices 693 16B.1 Binary Choice Model 741
16.2.5 Wald Hypothesis Tests 695 16B.2 Probit or Logit? 742
16.2.6 Likelihood Ratio Hypothesis Tests 696
Appendix 16C Using Latent Variables 743
16.2.7 Robust Inference in Probit and Logit
16C.1 Tobit (Tobit Type I) 743
Models 698
16C.2 Heckit (Tobit Type II) 744
16.2.8 Binary Choice Models with a Continuous
Endogenous Variable 698 Appendix 16D A Tobit Monte Carlo Experiment 745

k
 k

Table of Contents xix

A Mathematical Tools 748

B.2.5 Distributions of Functions of Random
Variables 787
B.2.6 Truncated Random Variables 789
A.1 Some Basics 749
A.1.1 Numbers 749 B.3 Some Important Probability
A.1.2 Exponents 749 Distributions 789
A.1.3 Scientific Notation 749 B.3.1 The Bernoulli Distribution 790
A.1.4 Logarithms and the Number e 750 B.3.2 The Binomial Distribution 790
A.1.5 Decimals and Percentages 751 B.3.3 The Poisson Distribution 791
A.1.6 Logarithms and Percentages 751 B.3.4 The Uniform Distribution 792
B.3.5 The Normal Distribution 793
A.2 Linear Relationships 752
B.3.6 The Chi-Square Distribution 794
A.2.1 Slopes and Derivatives 753
B.3.7 The t-Distribution 796
A.2.2 Elasticity 753
B.3.8 The F-Distribution 797
A.3 Nonlinear Relationships 753 B.3.9 The Log-Normal Distribution 799
A.3.1 Rules for Derivatives 754
B.4 Random Numbers 800
A.3.2 Elasticity of a Nonlinear Relationship 757
B.4.1 Uniform Random Numbers 805
A.3.3 Second Derivatives 757
A.3.4 Maxima and Minima 758 B.5 Exercises 806
A.3.5 Partial Derivatives 759
A.3.6 Maxima and Minima of Bivariate
Functions 760
C Review of Statistical
A.4 Integrals 762
Inference 812
A.4.1 Computing the Area Under a
C.1 A Sample of Data 813
Curve 762
C.2 An Econometric Model 814
k A.5 Exercises 764 k
C.3 Estimating the Mean of a Population 815
C.3.1 The Expected Value of Y 816
B Probability Concepts 768 C.3.2 The Variance of Y 817
C.3.3 The Sampling Distribution of Y 817
B.1 Discrete Random Variables 769 C.3.4 The Central Limit Theorem 818
B.1.1 Expected Value of a Discrete Random C.3.5 Best Linear Unbiased Estimation 820
Variable 769
B.1.2 Variance of a Discrete Random C.4 Estimating the Population Variance and Other
Variable 770 Moments 820
B.1.3 Joint, Marginal, and Conditional C.4.1 Estimating the Population
Distributions 771 Variance 821
B.1.4 Expectations Involving Several Random C.4.2 Estimating Higher Moments 821
Variables 772 C.5 Interval Estimation 822
B.1.5 Covariance and Correlation 773 C.5.1 Interval Estimation: σ2 Known 822
B.1.6 Conditional Expectations 774 C.5.2 Interval Estimation: σ2 Unknown 825
B.1.7 Iterated Expectations 774
C.6 Hypothesis Tests About a Population
B.1.8 Variance Decomposition 774
Mean 826
B.1.9 Covariance Decomposition 777
C.6.1 Components of Hypothesis Tests 826
B.2 Working with Continuous Random C.6.2 One-Tail Tests with Alternative ‘‘Greater
Variables 778 Than’’ (>) 828
B.2.1 Probability Calculations 779 C.6.3 One-Tail Tests with Alternative ‘‘Less
B.2.2 Properties of Continuous Random Than’’ (<) 829
Variables 780 C.6.4 Two-Tail Tests with Alternative ‘‘Not Equal
B.2.3 Joint, Marginal, and Conditional Probability To’’ (≠) 829
Distributions 781 C.6.5 The p-Value 831
B.2.4 Using Iterated Expectations with Continuous C.6.6 A Comment on Stating Null and Alternative
Random Variables 785 Hypotheses 832

k
 k

xx Table of Contents

C.6.7 Type I and Type II Errors 833 C.10 Kernel Density Estimator 851
C.6.8 A Relationship Between Hypothesis Testing
C.11 Exercises 854
and Confidence Intervals 833
C.11.1 Problems 854
C.7 Some Other Useful Tests 834 C.11.2 Computer Exercises 857
C.7.1 Testing the Population Variance 834
C.7.2 Testing the Equality of Two Population
Means 834 D Statistical Tables 862
C.7.3 Testing the Ratio of Two Population
Variances 835 TableD.1 Cumulative Probabilities for the Standard
C.7.4 Testing the Normality of a Population 836 Normal Distribution 𝚽(z) = P(Z ≤ z) 862
C.8 Introduction to Maximum Likelihood TableD.2 Percentiles of the t-distribution 863
Estimation 837
TableD.3 Percentiles of the Chi-square
C.8.1 Inference with Maximum Likelihood
Distribution 864
Estimators 840
C.8.2 The Variance of the Maximum Likelihood TableD.4 95th Percentile for the F-distribution 865
Estimator 841 TableD.5 99th Percentile for the F-distribution 866
C.8.3 The Distribution of the Sample
TableD.6 Standard Normal pdf Values 𝛟(z) 867
Proportion 842
C.8.4 Asymptotic Test Procedures 843
C.9 Algebraic Supplements 848 INDEX 869
C.9.1 Derivation of Least Squares Estimator 848
C.9.2 Best Linear Unbiased Estimation 849

k k