Flexible Regression Models for Count Data Based on Renewal Processes: The Countr Package

Abstract

A new alternative to the standard Poisson regression model for count data is suggested. This new family of models is based on discrete distributions derived from renewal processes, i.e., distributions of the number of events by some time t. Unlike the Poisson model, these models have, in general, time-dependent hazard functions. Any survival distribution can be used to describe the inter-arrival times between events, which gives a rich class of count processes with great flexibility for modelling both underdispersed and overdispersed data. The R package Countr provides a function, renewalCount(), for fitting renewal count regression models and methods for working with the fitted models. The interface is designed to mimic the glm() interface and standard methods for model exploration, diagnosis and prediction are implemented. Package Countr implements stateof-the-art recently developed methods for fast computation of the count probabilities. The package functionalities are illustrated using several datasets.

Comparison of Chronic Wound Inpatients and Outpatients’ Diets and Meals Nutrient Content in Taabo Wound Management Unit, Côte d’Ivoire

“…The data entered in Kobo collect was transferred and processed with Microsoft Excel [20] to check for missing values. The final database was exported to in the R data analysis software to be analyzed and visualized with the basic function "glm" and the package "lmtest" [21]. Multiple match analysis (MCA) was per- formed to summarize the data and allow multiple categorical variables to be visualized in a single dimension (the Kruskal-Wallis's test was performed for more than two groups and the significance level was set at P < 0.05) [22].…”

PoARX Modelling for Multivariate Count Time Series

“…Another direction would be to consider distributions other than Poisson. We are considering the possibility of using the renewal count distributions of Kharrat et al (2018), mentioned in the introduction, which are implemented in the R package Rcountr (Kharrat and Boshnakov, 2016). Combining these renewal distributions with the ideas found in this paper could lead to a fascinating new family of count time series models.…”

“…McShane et al (2008) developed a count model based on the Weibull distribution that can handle both under-dispersed and overdispersed data. Building on this idea, Kharrat et al (2018) extended this approach to create a rich and flexible family of renewal count distributions, which greatly extends the toolbox of distributions available for modelling count data.…”

A flexible Bayesian nonconfounding spatial model for analysis of dispersed count data

“…(2019) extended the GC model for analysis of spatially correlated count data. Furthermore, some R packages such as countr (Kharrat et al., 2019) have been introduced for implementing the idea in a nonspatial framework.…”