Accounting for Excess Zeros and Sample Selection in Poisson and Negative Binomial Regression Models

Source: RePEc

ABSTRACT We present several modifications of the Poisson and negative binomial models for count data to accommodate cases in which the number of zeros in the data exceed what would typically be predicted by either model. The excess zeros can masquerade as overdispersion. We present a new test procedure for distinguishing between zero inflation and overdispersion. We also develop a model for sample selection which is analogous to the Heckman style specification for continuous choice models. An application is presented to a data set on consumer loan behavior in which both of these phenomena are clearly present.

  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: The e¤ect of public …nancing on health care utilization is analyzed based on various empirical speci…-cations. Inpatient care, and utilization of outpatient care provided by general practitioners and specialists are considered. Two-stage utilization decisions, and the endogeneity of private health insurance are taken into account in the empirical models. The empirical …ndings indicate positive e¤ect of public …nancing on health care utilization, which is modest in absolute terms but not negligible in relative terms. There is no clear evidence for an increasing e¤ect of supplementary private health insurance on utilization. The results are based on the …rst wave of the Survey of Health, Ageing and Retirement in Europe, a cross-national micro database of individuals aged 50 or over.
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Innovation networks can contribute both to the creation of new knowledge and to the diffu-sion of existing and newly created knowledge. The Industrial Co-operative Research (ICR) programme aims at both targets at the same time. The success of this programme depends essentially on the performance of the innovation networks which have developed in close as-sociation with public financing. This contribution presents results obtained from evaluating network activities at the project level. The participation of firms in several network activities and the patterns of adoption of the new knowledge generated are analysed. By applying probit and zero-inflated Poisson regressions, we find that closeness of cooperation between research institutes and firms both in the creation and diffusion of knowledge is related with the subse-quent use of project results.
  • [Show abstract] [Hide abstract]
    ABSTRACT: The study of proportions is a common topic in many fields of study. The standard beta distribution or the inflated beta distribution may be a reasonable choice to fit a proportion in most situations. However, they do not fit well variables that do not assume values in the open interval (0,c),0<c<1. For these variables, the authors introduce the truncated inflated beta distribution (TBEINF). This proposed distribution is a mixture of the beta distribution bounded in the open interval (c,1) and the trinomial distribution. The authors present the moments of the distribution, its scoring vector, and the Fisher information matrix, and discuss estimation of its parameters. The properties of the suggested estimators are studied using Monte Carlo simulations. In addition, the authors present an application of the TBEINF distribution for unemployment insurance data.
    Communication in Statistics- Theory and Methods 03/2012; · 0.28 Impact Factor

Full-text (2 Sources)

Available from
Jun 2, 2014