Econometric analysis of count data.
This second, revised and enlarged edition, for the review of the first edition from 1994 see Zbl 0810.90021, provides an excellent overview of the current state of the art in the analysis of event counts. It is just a decade ago when econometricians started to use this technique in applied empirical work such as labour economics, urban economics and economic demography. The author starts with an introduction of probability models for event counts. Several models for event histories are presented (individual event history data, univariate Poisson processes, dynamic probability models) and different distributions for count data are discussed. Among these are the Poisson distribution, die binomial-, the negative binomial- and the logarithmic distribution. A special subsection is devoted to the problem of truncated count data distributions. Besides of these basic distributions the author also presents some generalizations as far as possible. A final section deals with duration analysis and count data. After this ``statistical view, the author directs interest on econometric modeling. The basic issues consist of the Poisson regression model, potential sources of misspecification and its consequences for the estimates. Because misspecification is one of the most serious problems in the analysis of count data, several tests of misspecification are provided by the author. Having introduced the basics, the next chapter deals with possible extensions how to proceed when unobserved heterogeneity is present or when selectivity occurs exogenously or endogenously. Another extension relates to generalized variance models including topics such as robust Poisson regression as well as nonparametric variance estimation. Although most of the economic applications are microeconometric in nature, count data models are also applicable for time series of counts. This issue is addressed in the chapter ``Further Topics. Included in this chapter is also a brief introduction to the Bayesian analysis of count variables. The basic issues of the book are demonstrated with an application to labour mobility. Although primarily written for applied econometricians the monograph is also useful to statisticians and biometricians as well.
- A count data model with unobserved heterogeneity
- A regression model for overdispersed data without too many zeros
- Regression modeling of one-inflated positive count data
- Inferences in semi-parametric dynamic mixed models for longitudinal count data
- Three estimators for the Poisson regression model with measurement errors
- Two stage cluster sampling based asymptotic inferences in survey population models for longitudinal count and categorical data
- Analyzing unevenly spaced longitudinal count data
- Overdispersed and underdispersed Poisson generalizations
- A comparison of asymptotic covariance matrices of three consistent estimators in the Poisson regression model with measurement errors
- Bayesian-type count data models with varying coefficients: estimation and testing in the presence of overdispersion
- Point and interval estimation of the population size using the truncated Poisson regression model
- scientific article; zbMATH DE number 1237531 (Why is no real title available?)
- Count data models with selectivity
- Count data models for demographic data∗
- A note on modelling underreported Poisson counts
- scientific article; zbMATH DE number 2002520 (Why is no real title available?)
- GENERALIZED INTEGER-VALUED AUTOREGRESSION
- scientific article; zbMATH DE number 1495545 (Why is no real title available?)
- Longitudinal modeling of insurance claim counts using jitters
- Regression analysis of count data
- Event count estimation
- Unusual-event processes for count data
- Financial data modeling by Poisson mixture regression
- On the fast computation of the Dirichlet-multinomial log-likelihood function
- The impact of covariate measurement error on risk prediction
- Non nested model selection for spatial count regression models with application to health insurance
- Bootstrap-based model selection criteria for beta regressions
This page was built for publication: Econometric analysis of count data.
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4337695)
