A two-tail version of the PPS distribution with application to current account balance data

Abstract

The double Pareto-positive stable (dPPS) distribution is introduced as a new model for describing countries’ global current account balance data. The dPPS distribution provides a flexible model for fitting the entire range of a set of current account data (both surplus and deficit), where zero and unimodality are possible, and the double-sided Pareto distribution is included as a particular case. Expressions for the cumulative distribution, probability density and quantile functions are given. An estimation method is discussed and a simple graphical method for studying the adequacy of the data to model is given. Finally, we consider the fit of countries’ global current account balance data for several years. The new distribution is compared with four classical models: Normal, Skew Normal, Asymmetric Laplace and Hyperbolic distributions. In all the data sets considered, the dPPS distribution outperforms the fits of these four distributions.