INSEAD Library

Tanoto copy is hardback

Includes bibliographical references and index

Digitized

Discrete Choice Methods with Simulation Contents 1 Introduction 1.1 Motivation 1.2 Choice Probabilities and Integration 1.3 Outline of Book 1.4 A Couple of Notes Part I Behavioral Models 2 Properties of Discrete Choice Models 2.1 Overview 2.2 The Choice Set 2.3 Derivation of Choice Probabilities 2.4 Specific Models 2.5 Identification of Choice Models 2.6 Aggregation 2.7 Forecasting 2.8 Recalibration of Constants 3 Logit 3.1 Choice Probabilities 3.2 The Scale Parameter 3.3 Power and Limitations of Logit 3.4 Nonlinear Representative Utility 3.5 Consumer Surplus 3.6 Derivatives and Elasticities 3.7 Estimation 3.8 Goodness of Fit and Hypothesis Testing 3.9 Case Study: Forecasting for a New Transit System 3.10 Derivation of Logit Probabilities 4 GEV 4.1 Introduction 4.2 Nested Logit page 1 1 3 7 8 11 11 11 14 17 19 29 32 33 34 34 40 42 52 55 57 60 67 71 74 76 76 77 4.3 Three-Level Nested Logit 4.4 Overlapping Nests 4.5 Heteroskedastic Logit 4.6 The GEV Family 5 Probit 5.1 Choice Probabilities 5.2 Identification 5.3 Taste Variation 5.4 Substitution Patterns and Failure of IIA 5.5 Panel Data 5.6 Simulation of the Choice Probabilities 6 Mixed Logit 6.1 Choice Probabilities 6.2 Random Coefficients 6.3 Error Components 6.4 Substitution Patterns 6.5 Approximation to Any Random Utility Model 6.6 Simulation 6.7 Panel Data 6.8 Case Study 7 Variations on a Theme 7.1 Introduction 7.2 Stated-Preference and Revealed-Preference Data 7.3 Ranked Data 7.4 Ordered Responses 7.5 Contingent Valuation 7.6 Mixed Models 7.7 Dynamic Optimization Part II Estimation 8 Numerical Maximization 8.1 Motivation 8.2 Notation 8.3 Algorithms 8.4 Convergence Criterion 8.5 Local versus Global Maximum 8.6 Variance of the Estimates 8.7 Information Identity 86 89 92 93 97 97 100 106 108 110 114 134 134 137 139 141 141 144 145 147 151 151 152 156 159 164 166 169 185 185 185 187 198 199 200 202 9 Drawing from Densities 9.1 Introduction 9.2 Random Draws 9.3 Variance Reduction 10 Simulation-Assisted Estimation 10.1 Motivation 10.2 Definition of Estimators 10.3 The Central Limit Theorem 10.4 Properties of Traditional Estimators 10.5 Properties of Simulation-Based Estimators 10.6 Numerical Solution 11 Individual-Level Parameters 11.1 Introduction 11.2 Derivation of Conditional Distribution 11.3 Implications of Estimation of 9 11.4 Monte Carlo Illustration 11.5 Average Conditional Distribution 11.6 Case Study: Choice of Energy Supplier 11.7 Discussion 12 Bayesian Procedures 12.1 Introduction 12.2 Overview of Bayesian Concepts 12.3 Simulation of the Posterior Mean 12.4 Drawing from the Posterior 12.5 Posteriors for the Mean and Variance of a Normal Distribution 12.6 Hierarchical Bayes for Mixed Logit 12.7 Case Study: Choice of Energy Supplier 12.8 Bayesian Procedures for Probit Models 13 Endogeneity 13.1 Overview 13.2 The BLP Approach 13.3 Supply Side 13.4 Control Functions 13.5 Maximum Likelihood Approach 13.6 Case Study: Consumers' Choice among New Vehicles 205 205 205 214 237 237 238 245 247 250 257 259 259 262 264 267 269 270 280 282 282 284 291 293 294 299 305 313 315 315 318 328 334 340 342 14 EM Algorithms 14.1 Introduction 14.2 General Procedure 14.3 Examples of EM Algorithms 14.4 Case Study: Demand for Hydrogen Cars Bibliography Index 347 347 348 355 365 371 385