The “count data” ?eld has further ?ourished since the previous edition of this book was published in 2003. The development of new methods has not slowed down by any means, and the application of existing ones in applied work has expanded in many areas of social science research. This, in itself, would be reason enough for updating the material in this book, to ensure that it continues to provide a fair representation of the current state of research. In addition, however, I have seized the opportunity to undertake some major changes to the organization of the book itself. The core material on cross-section models for count data is now presented in four chapters, rather than in two as previously. The ?rst of these four chapters introduces the Poissonregressionmodel,anditsestimationbymaximumlikelihoodorpseudo maximum likelihood. The second focuses on unobserved heterogeneity, the third on endogeneity and non-random sample selection. The fourth chapter provides an extended and uni?ed discussion of zeros in count data models. This topic deserves, in my view, special emphasis, as it relates to aspects of modeling and estimation that are speci?c to counts, as opposed to general exponential regression models for non-negative dependent variables. Count distributions put positive probability mass on single o- comes, and thus o?er a richer set of interesting inferences.
The book provides graduate students and researchers with an up-to-date survey of statistical and econometric techniques for the analysis of count data, with a focus on conditional distribution models. Proper count data probability models allow for rich inferences, both with respect to the stochastic count process that generated the data, and with respect to predicting the distribution of outcomes. The book starts with a presentation of the benchmark Poisson regression model. Alternative models address unobserved heterogeneity, state dependence, selectivity, endogeneity, underreporting, and clustered sampling. Testing and estimation is discussed from frequentist and Bayesian perspectives. Finally, applications are reviewed in fields such as economics, marketing, sociology, demography, and health sciences. The fifth edition contains several new topics, including copula functions, Poisson regression for non-counts, additional semi-parametric methods, and discrete factor models. Other sections have been reorganized, rewritten, and extended.
