Contents
List of Tables xi
List of Figures xiv
Series Foreword xvii
Preface xix
1
Introduction
1
1.1 The Context of Transportation Demand Forecasting 1
1.2 The Background of Discrete Choice Analysis 2
1.3 Transportation Applications of Discrete Choice Analysis 3
1.4 Outline of the Book 4
2
Review of the Statistics of Model Estimation
7
2.1 The Estimation Problem 7
2.2 Criteria for Evaluating Estimators 11
2.3 Small Sample Properties 12
2.4 Asymptotic Properties 17
2.5 Methods of Estimation 20
2.6 Key Statistical Tests 23
2.7 Summary 29
3
Theories of Individual Choice Behavior
31
3.1 Introduction 31
3.2 A Framework for Choice Theories 31
3.3 Rational Behavior 38
3.4 Economic Consumer Theory 39
3.5 Extensions of Consumer Theory 42
3.6 Discrete Choice Theory 43
3.7 Probabilistic Choice Theory 48
3.8 Summary 57
4
Binary Choice Models
59
4.1 Making Random Utility Theory Operational 60
4.2 Common Binary Choice Models 66
4.3 Examples of Binary Choice Models 74
Page viii
4.4 Maximum Likelihood Estimation of Binary Choice Models 79
4.5 Examples of Maximum Likelihood Estimation 87
4.6 Other Estimation Methods for Binary Choice Models 94
4.7 Summary 99
5
Multinomial Choice
100
5.1 Theory of Multinomial Choice 100
5.2 The Multinomial Logit Model 103
5.3 Properties of Logit 108
5.4 Specification of a Multinomial Logit Model 114
5.5 Estimation of Multinomial Logit 118
5.6 Example of Estimation Results 121
5.7 Other Multinomial Choice Models 123
5.8 Summary 129
6
Aggregate Forecasting Techniques
131
6.1 The Problem of Aggregation across Individuals 131
6.2 Typology of Aggregation Methods 134
6.3 Description of Aggregation Procedures 135
6.4 A Comparison of Methods for Aggregate Forecasting 148
6.5 Summary 153
7
Tests and Practical Issues in Developing Discrete Choice Models
154
7.1 Introduction 154
7.2 The Art of Model Building 154
7.3 A Mode Choice Model Example 155
7.4 Tests of Alternative Specifications of Variables 157
7.5 Tests of the Model Structure 183
7.6 Prediction Tests 207
7.7 Summary 216
8
Theory of Sampling
217
8.1 Basic Sampling Concepts 218
8.2 Overview of Common Sampling Strategies 221
Page ix
8.3 Sampling Strategies for Discrete Choice Analysis 229
8.4 Estimating Choice Models under Alternative Sampling Strategies 234
8.5 Choosing a Sample Design for Discrete Choice Analysis 244
8.6 Summary 250
9
Aggregation and Sampling of Alternatives
253
9.1 Introduction 253
9.2 Aggregation of Alternatives 253
9.3 Estimation of Choice Models with a Sample of Alternatives 261
9.4 Estimation Results for Three Destination Choice Models 269
9.5 Summary 275
10
Models of Multidimensional Choice and the Nested Logit Model
276
10.1 Multidimensional Choice Sets 276
10.2 Multidimensional Choice Sets with Shared Observed Attributes:
Joint Logit
278
10.3 Multidimensional Choice Models with Shared Unobserved
Attributes: Nested Logit
285
10.4 Estimating the Nested Logit Model 295
10.5 Multidimensional Choice Models with Shared Unobserved
Attributes: Multinomial Probit
299
10.6 Measure of Accessibility 300
10.7 Derivation of the Nested Logit Model from the Generalized
Extreme Value Model
304
10.8 An Example of a Multidimensional Choice Model 310
10.9 Summary 321
11
Systems of Models
323
11.1 Introduction 323
11.2 Issues in Model System Design 324
11.3 A System of Urban Travel Demand Models for Metropolitan
Region Transportation Planning
327
11.4 A Short-Range Travel Demand Model System 338
11.5 Summary 357
12
Models of Travel Demand: Future Directions
359
12.1 Introduction 359
12.2 Components of Travel Demand Modeling Process 359
12.3 Behavioral Theory 360
12.4 Measurement 364
12.5 Statistical Model Structure 368
12.6 Estimation 371
12.7 Summary 372
Bibliography 374
Index 385