This book is a treatise on empirical microeconomics: it describes the econometric theory of qualitative choice models and the empirical practice of modeling consumer demand for a heterogeneous commodity, housing. Accordingly, the book has two parts. The first part gives a self-contained survey of discrete choice models with emphasis on nested and related multinomial logit models. The second part concentrates on three sUbstantive questions about housing demand and how they can be answered using discrete choice models. Why combine these two distinct parts in one book? It is the interaction between theory and application in empirical microeconomics on which we focus in this book. Hence, emphasis in the methodological part is on practicability, and emphasis in the applied part is on the usage of the proper econometric specifications. Econometrics means measuring economic phenomena. Because nature (ironically, in the case of economics, this is most often the government) rarely provides us with well-defined economic experiments, measurement of economic phenomena usually requires an elaborate statistical apparatus that is able to separate concurrent and confounding phenomena. Discrete choice models have proved to be a very convenient apparatus to study the complex issues in housing demand. We present models, techniques, and statistical problems of discrete choice in the first and methodological part of the book, written in conventional textbook style.

Inhaltsangabe1: Introduction.- 1.1 Substantive Issues.- 1.2 Methodological Issues.- 1.3 Organization of the Book.- One: Econometric Foundations Discrete Choice Analysis.- 2: The Random utility Maximization Hypothesis.- 2.1 Continuous Versus Discrete Choice.- 2.2 Starting from the Utility Function.- 2.3 Starting from Choice Probabilities.- 3: Functional Specification of Discrete Choice Models.- 3.1 The Probit Model.- 3.2 The Logit Model.- 3.3 Extensions of the Logit Approach.- 3.4 The Linear Probability Model.- 3.5 Nonparametric Specifications.- 4: The Nested Multinominal Logit Model.- 4.1 Hierarchical Choice.- 4.2 Relation to the Random Utility Maximization Hypothesis.- 4.3 Estimation by Maximum Likelihood.- 4.4 Specification Tests.- 4.5 Summary.- 5: Panel Data.- 5.1 Discrete Choice for Pooled Cross-Sectional Data.- 5.2 Fixed Effects.- 5.3 Random Effects.- 5.4 Fixed Effects versus Random Effects Specifications.- 6: Economical Sampling and Estimation Techniques.- 6.1 Random Sampling of Alternatives.- 6.2 Choice Based Sampling.- 6.3 Fitting Aggregate Probability Shares.- 6.4 Estimation with Grouped Data.- 6.5 Goodness-of-Fit Measures.- Two: Applications: The Demand for Housing in the United States and West Germany.- 7: Housing Choices.- 7.1 Discrete Choice Description of Housing Demand.- 7.2 Explanatory Variables.- 7.3 Data Sources.- 8: Housing Preferences in the United States and West Germany.- 8.1 Introduction.- 8.2 Specification of Price Variables and Hedonic Regression.- 8.3 Specification of Income Variables and Permanent Income Estimation.- 8.4 Specification of Demographic Variables.- 8.5 Specification of Nesting Structures.- 8.6 Empirical Results.- 8.7 Conclusions.- 9: The Household Formation Decision.- 9.1 Introduction.- 9.2 Background and Nucleus Decomposition.- 9.3 Specification of the Demand Equations.- 9.4 Empirical Results.- 9.5 Conclusions.- 10: Tracing Housing Choices Over Time.- 10.1 Introduction.- 10.2 Data and Specification of the Demand Equations.- 10.3 Comparison: Panel Data and Cross-Sectional Analysis.- 10.4 Conclusions.