Modelagem probabilística de dados de pagamentos de provedor de internet usando variável mista

Abstract

The objective of this work is to use a mixed random variable to model data from payments made by customers of a Paraíba city. In a data analysis, one of the first steps is to observe the nature of the variables involved and to make a graphical analysis of them. Generally, these variables can be classified as discrete or continuous. The discrete ones arise predominantly from categorizations or countings, while the continuous ones arise from measures. But there are still the mixed variables, which are obtained by making a weighted sum of discrete and continuous variables. In the case of the data used here, the graphical analysis indicated an exponential behavior, which is a continuous model for positive data. However, there was a large amount of null values, which gave rise to the idea of using a mixed variable, and the positive part will be modeled by the exponential distribution and the null ones by a degenerate variable at zero point.