Bayesvl: an R package for user-friendly Bayesian regression modelling

Authors

  • Quan-Hoang Vuong
  • Minh-Hoang Nguyen
  • Manh-Toan Ho*

Keywords:

Bayesian network, bayesvl, ggplot2, mathematical model, MCMC

Abstract

Compared with traditional statistics, only a few social scientists employ Bayesian analyses. The existing software programs for implementing Bayesian analyses such as OpenBUGS, WinBUGS, JAGS, and rstanarm can be daunting given that their complex computer codes involve a steep learning curve. In contrast, this paper introduces a new open software for implementing Bayesian network modelling and analysis: the bayesvl R package. The package aims at providing an intuitive gateway for beginners of Bayesian statistics to construct and analyse mathematical models in social sciences. To achieve this aim, the bayesvl package integrates three core functions seamlessly: (i) designing Bayesian network models using directed acyclic graphs (DAGs) of bnlearn, (ii) generating attractive visualization of ggplot2, and (iii) simulating data and computing posterior distribution using the Markov chain Monte Carlo (MCMC) algorithms of rstan and rethinking. A case example illustrates how the bayesvl package helps leverage users’ intuition in creating and evaluating mathematical models of their social scientific problems while minimizing the daunting aspect of writing complex computer codes.

DOI:

https://doi.org/10.31276/VMOSTJOSSH.64(1).85-96

Classification number

7

Author Biographies

Quan-Hoang Vuong

Centre for Interdisciplinary Social Research, Phenikaa University, Hanoi, Vietnam

Vietnam Institute for Advanced Study in Mathematics, Ministry of Education and Training

Minh-Hoang Nguyen

Centre for Interdisciplinary Social Research, Phenikaa University, Hanoi, Vietnam

Manh-Toan Ho

Centre for Interdisciplinary Social Research, Phenikaa University, Hanoi, Vietnam

Downloads

Published

2022-04-20

Received 5 october 2021; accepted 23 February 2022

How to Cite

Quan-Hoang Vuong, Minh-Hoang Nguyen, & Manh-Toan Ho. (2022). Bayesvl: an R package for user-friendly Bayesian regression modelling. The VMOST Journal of Social Sciences and Humanities, 64(1), 85-96. https://doi.org/10.31276/VMOSTJOSSH.64(1).85-96

Issue

Section

Other Social Sciences

Most read articles by the same author(s)