Abstract

In recent years, statistical models and techniques for discrete data analysis have advanced significantly in a wide range of research areas. Even when the phenomenon of interest is theoretically continuous, but its observed values are registered as discrete (just think of the lifetime of many devices in industry, when one observes the number of cycles completed before failure), it is more appropriate to adopt a discrete model derived as an analog of the hypothesized continuous distribution, preserving one or more of its features. This session is open to all contributions on statistical analysis of discrete data, with a particular focus on discrete counterparts of univariate and multivariate continuous distributions and their reliability and dependence properties.