Title: BayesTwin - An R package for Bayesian Analysis of Twin Data

Authors: Inga Schwabe, Stˇphanie M. van den Berg

Affiliation: University of Twente

Abstract: In a behavior genetics study, one is typically interested in
determining the relative contributions of nature and nurture in causing
individual differences in a behavioral trait.  To this end, often twin data
is used to decompose observed variance in family members into genetic
variance, shared-environmental variance and unique-environmental variance.
Where in the last quarter of the 20th century, mostly structural equation
modeling (SEM) was used for statistical inference, the first decade of the
21st century has seen increased use of Bayesian modeling.  This has opened
up new modeling possibilities such as inference on complex models that were
not easily tractable using standard frequentist techniques, including
models based on Item Response Theory (IRT) or modeling gene-environment
interaction in the presence of gene-environment correlation.  However, the
twin research community still seems reluctant to embrace this new
technology, partly due to a lack of standard for reporting results based on
twin data.  To make Bayesian analysis more accessible for twin researchers,
we introduce the R package BayesTwin.  The package includes a wide range of
twin models that can be estimated using the program JAGS as well as
functions that determine whether the analysis was performed well and
functions to plot relevant information.  BayesTwin can be downloaded from
the GitHub account of the first author.