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.
Supplementary materials are available in addition to this article. It can be downloaded at RJ-2020-003.zip
mistr, distr, CompLognormal, evmix, OpVar, ReIns, gendist, ggplot2, actuar, bbmle
Distributions, ExtremeValue, Finance, Graphics, Phylogenetics, Robust, TeachingStatistics
Text and figures are licensed under Creative Commons Attribution CC BY 4.0. The figures that have been reused from other sources don't fall under this license and can be recognized by a note in their caption: "Figure from ...".
For attribution, please cite this work as
Sablica & Hornik, "The R Journal: mistr: A Computational Framework for Mixture and Composite Distributions", The R Journal, 2019
BibTeX citation
@article{RJ-2020-003,
author = {Sablica, Lukas and Hornik, Kurt},
title = {The R Journal: mistr: A Computational Framework for Mixture and Composite Distributions},
journal = {The R Journal},
year = {2019},
note = {https://doi.org/10.32614/RJ-2020-003},
doi = {10.32614/RJ-2020-003},
volume = {12},
issue = {1},
issn = {2073-4859},
pages = {283-299}
}