Advances and Applications in Statistics
Volume 6, Issue 3, Pages 305 - 322
(December 2006)
|
|
GENERALIZED POISSON REGRESSION MODEL FOR UNDERREPORTED COUNTS
Mavis Pararai (U. S. A.), Felix Famoye (U. S. A.) and Carl Lee (U. S. A.)
|
Abstract: The generalized Poisson regression model has been used to model equi-, over- and under-dispersed count data. In many of these situations the assumption is that the response, the count, is reported without error. It is possible that the count may be underreported or overreported. The Poisson regression model and the negative binomial regression model for underreported counts have been developed in the literature. In this paper, the generalized Poisson regression model for underreported counts is studied. The parameters of the proposed model are estimated by the maximum likelihood method. A score test to determine whether there is significant underreporting in the data is derived. Finally, a numerical example is used to illustrate the generalized Poisson regression model for underreported counts. |
Keywords and phrases: over- and under-dispersion, estimation, hypothesis testing, sex partner data. |
|
Number of Downloads: 360 | Number of Views: 1189 |
|