Modelling Brazilian Hyperinflation via Lomax-Tangent-Fisk Distribution

In recent years, many researchers have focused on the development of new probability distributions. The massive amount of data available in real life encourages the study of models which can provide more flexibility than the classical ones do. In this paper, we tackle the thorny problem related to the Brazilian inflation dynamics. Thus, we employ the IPCA, which is the primary Brazilian inflation index. To this extent, we perform a systematic statistical investigation, and we propose a novel extension of the well-known Fisk distribution called Lomax-Tangent-Fisk. We discuss some structural properties of the stochastic model, such as the quantile function, the linear representation of the probability density function, the moments and the moment generating function. We also fitted the proposed model to the IPCA data using the maximum likelihood method, and we compared it with the fits of other six distributions. Goodness-of-fit measures were used as criteria of comparison. The results attest that the proposed distribution outperforms the competing models according to the information criteria and proper goodness-of-fit statistics