Abstract
Understanding why a random variable is actually random has been in the core of Statistics from its beginnings. The generalized Waring regression model for count data explains that inherent variability is given by three possible sources: randomness, liability and proneness. The model extends the negative binomial regression model and it is not included in the family of generalized linear models. In order to avoid that shortcoming, we developed the GWRM R package for fitting, describing and validating the model. The version we introduce in this communication provides a new design of the modelling function as well as new methods operating on the associated fitted model objects, so that the new software integrates easily into the computational toolbox for modelling count data in R. The release of a plug-in in order to use the package from the interface R Commander tries to contribute to the spreading of the model among non-advanced users. We illustrate the usage and the possibilities of the software with two examples from the fields of health and sport.
Highlights
The apparently chaotic behaviour of any random variable reveals that, in general, we do not know how and why data vary
That is the origin of mixed Poisson distributions, which allows the occurrence rate to vary among different cases; in the end, they assume there are two sources of variability in the counting process, pure chance and differences between individuals
The Univariate Generalized Waring Distribution (UGWD) [4] allows for identifying a new possible source of variability. It may be considered as a mixture of a negative binomial (NB) distribution, in such a way that when a count variable follows a UGWD, it may be inferred that the variable is affected by three possible sources of variation: pure chance, PLOS ONE | DOI:10.1371/journal.pone
Summary
GWRM: An R Package for Identifying Sources of Variation in Overdispersed Count Data. The generalized Waring regression model for count data explains that inherent variability is given by three possible sources: randomness, liability and proneness. In order to avoid that shortcoming, we developed the GWRM R package for fitting, describing and validating the model. The version we introduce in this communication provides a new design of the modelling function as well as new methods operating on the associated fitted model objects, so that the new software integrates into the computational toolbox for modelling count data in R. The release of a plug-in in order to use the package from the interface R Commander tries to contribute to the spreading of the model among non-advanced users. We illustrate the usage and the possibilities of the software with two examples from the fields of health and sport
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
