The R Journal: accepted article

This article will be copy edited and may be changed before publication.

mistr: A Computational Framework for Mixture and Composite Distributions PDF download
Lukas Sablica and Kurt Hornik

Abstract Finite mixtures and composite distributions allow to model the probabilistic representation of data with more generality than simple distributions and are useful to consider in a wide range of applications. The R package mistr provides an extensible computational framework for creating, transforming, and evaluating these models, together with multiple methods for their visualization and description. In this paper we present the main computational framework of the package and illustrate its application. In addition, we provide and show functions for data modeling using two specific composite distributions as well as a numerical example where a composite distribution is estimated to describe the log-returns of selected stocks.

Received: ; online 2019-12-27, supplementary material, (1.9 Kb)
CRAN packages: mistr, distr, CompLognormal, evmix, OpVar, ReIns, gendist, ggplot2, actuar, bbmle
CRAN Task Views implied by cited CRAN packages: Distributions, ExtremeValue, Finance, Graphics, Phylogenetics, Robust, TeachingStatistics


CC BY 4.0
This article and supplementary materials are licensed under a Creative Commons Attribution 4.0 International license.

@article{RJ-2020-003,
  author = {Lukas Sablica and Kurt Hornik},
  title = {{mistr: A Computational Framework for Mixture and Composite
          Distributions}},
  year = {2020},
  journal = {{The R Journal}},
  doi = {10.32614/RJ-2020-003},
  url = {https://journal.r-project.org/archive/2020/RJ-2020-003/index.html}
}