Changes in version 1.5.4 o option to use "rjags" package for JAGS sampling via method = "rjags" in traitMPT/betaMPT. Changes in version 1.5.3 (2025-09-23) o Update for RcppArmadillo 15 Changes in version 1.5.2 o Bugfix for issue #15 by @singmann: function readEQN() no longer ignores the given restrictions but returns two columns ("Equation" without restrictions and "EQN" with restrictions) Changes in version 1.5.1 o bug fix for numerical instability of Bayes-factor computations when all models do not fit o allow formula syntax as in: predStructure = list("D ~ x1 + x2") Changes in version 1.5.0 (2023-05-21) o new argument monitorIndividual = TRUE for traitMPT() and betaMPT() allows to disable storing MCMC samples for individual theta parameters o R code cleaned up with package "styler" o bug fix for issue #12 by @soelderer (Paul): plotConvergence() uses strange parameter labels for models with more than 10 parameters o bug fix for issue #11 by @mariusbarth: issues with printing of WAIC() results Changes in version 1.4.9 (2022-09-06) o fix deprecated << operator in RcppArmadillo o added references to MPT tutorial (Schmidt et al., 2023) Changes in version 1.4.8 (2022-08-23) o check issue with HTML5 code (align, hspace) o improved documentation of the probit regression in traitMPT() o bugfix for getSamples() by @mariusbarth (GitHub issue #9) Changes in version 1.4.7 (2021-01-08) o bugfix for computing transformedParameters(..) for more than two parameters at the individual level Changes in version 1.4.6 o Minor bugfix for naming of categories in plotFit() function o Minor bugfix for BayesFactorMPT() Changes in version 1.4.5 (2020-05-31) o added standardized regression slopes for latent-trait MPT regression with continuous predictors (i.e., traitMPT with the option predStructure) o BayesFactorSlope() plots the posterior distribution of the unstandardized (instead of partially standardized) slope parameter o bugfix: posterior predictive check (PPP) for MPT models with a single tree Changes in version 1.4.4 (2019-12-05) o Improvements and bugfixes for the C++ MCMC samplers by Marius Barth. o Bug fixes for issues concerning class(matrix(...)) in R 4.0.0 Changes in version 1.4.3 (2019-04-04) o Thinning within C++ samplers [thanks to Marius Barth] o New default priors for alpha and beta in betaMPT: "dgamma(1,.1)T(1,)" Changes in version 1.4.2 o License updated to GPL-3 o Standard error for WAIC o Hexagon sticker added Changes in version 1.4.1 (2018-12-18) o summarizeMCMC: Batchwise computation of summary statistics to reduce RAM overload o WAIC implemented (based on a developmental feature of JAGS) Changes in version 1.3.2 o Bugfix: extendMPT() for betaMPT o New function to extract MCMC samples: getSamples(...) o Possibility to extract MCMC samples for getGroupMeans(...) Changes in version 1.3.1 (2018-02-23) o Improved BayesFactorSlope(): JZS prior (=Cauchy) or g-prior(=normal) o Bugfixes due to uncorrelated traitMPT Changes in version 1.3.0 (2018-02-05) o New option for the latent-trait MPT model: Independent Cauchy priors instead of a multivariate Wishart prior for the random-effects covariance matrix. Changes in version 1.2.0 (2018-01-19) o new function BayesFactorSlope() to get Bayes factor for continuous predictors in traitMPT() o G^2 in PPP() o option to use mu instead of mean in genTraitMPT() o Various bugfixes Changes in version 1.1.1 (2017-08-23) o Various bug fixes Changes in version 1.1.0 (2017-04-01) o Functions to compute Bayes factors for simple (nonhierarchical, fixed-effects) MPT models: BayesFactorMPT() and marginalMPT() o Registration of C++ routines Changes in version 1.0.3 o Bugfix in readEQN: treat numeric parameter labels as constants o Bugfix for traitMPT in priorPredictive() o Optional arguments (...) for plotting functions Changes in version 1.0.0 o New function correlationPosterior() to estimate the posterior for the population correlation (taking into account the number of participants) o Second CRAN release Changes in version 0.8.2 o New function probitInverse() to get the bivariate transformation of mean and SD in probability space given a normal distribution in probit space o Visible function summarizeMCMC() to provide TreeBUGS-specific MCMC summaries o priorPredictive() allows to sample group-level parameters Changes in version 0.8.1 o Zeros-trick by Smith & Batchelder (2010) for betaMPT() implemented (i.e., by using the argument alpha="zero") o Parameter constraints result in the same parameter labels for JAGS and C++ samplers (e.g., "b=a=f" will be labeled "b") Changes in version 0.8.0 o New function withinSubjectsEQN() that replicates an MPT model multiple times with different tree, category, and parameter labels for within-subject factorial designs o plotFreq() uses boxplots instead of lines by default o Posterior-predictive p-values per participant o Bugfix: avoid negative "sigma" and reversed "rho" estimates in traitMPT() due to negative scaling parameters "xi" o Bugfix: posteriorPredictive() always sampled data for new participant (=> false T1+T2 posterior predictive checks!) Changes in version 0.7.1 o New function priorPredictive() to sample data sets from the prior o New function plotPrior() to plot prior distributions for group mean, SD, and correlations o Posterior predictive samples for new participants o Possible to use boxplot in plotFreq() o Improved data-generating functions to allow for parameter restrictions o Package testing via 'testthat' added Changes in version 0.7.0 o Possibility to define models directly in R without requiring EQN files (by providing the EQN equations in a character value) o New function transformedParameters() to obtain posterior samples of transformed parameters for a fitted model (both on the group and individual level) o Argument "transformedParameters" now accepts a path to a text file with transformations (one per line) o New argument "posteriorFile" to save fitted model and posterior MCMC samples in RData-file o Better compatibility of betaMPTcpp() results with plotting functions o Option to plot median or mean estimates in plotParam() o Increased stability of extendMPT() o Fixed bug in plotFreq() if input is given by a matrix/data.frame Changes in version 0.6.2 o Bugfix: betweenSubjects with different number of chains o Added data "arnold2013" by Arnold, Bayen, Kuhlmann, and Vaterrodt (2013) o More examples in R help files o Convergence plots for simpleMPT Changes in version 0.6.1 (2016-09-02) o First CRAN release o Fast C++ Gibbs sampler tailored standard (simple fixed-effects) MPT models (simpleMPT) o Adjusted summary functions to account for standard MPT models Changes in version 0.6.0 o Fast C++ Gibbs sampler tailored to beta-MPT models (betaMPTcpp) Changes in version 0.5.3 o Better weakly-informative priors for continuous predictors on the standardized scale Changes in version 0.5.2 o New function betweenSubjectMPT() that computes between-subject comparisons (e.g., differences) of parameters from two fitted hierarchical models o Extended plotFit() to fit observed against posterior-predicted covariances o Hyperpriors matched by vector names o Split vignette into "Intro" and "Extended" Changes in version 0.5.1 o Tests for participant/person heterogeneity implemented (Smith & Batchelder, 2008) Changes in version 0.5.0 o function fitModel() that fits both trait and beta MPT (avoids duplicated code) o correlations of parameters are computed in R, not in JAGS anymore (increase in speed) o argument "covStructure" removed (all correlations computed by default) o save posterior predictive samples in traitMPT/betaMPT object (i.e., expected/predicted/observed frequencies) o to get posterior predictive p-values, use ppp=1000 (previously: M.T1=1000) o getParam() and getGroupMean() allow to save results in .csv-file Changes in version 0.4.9 o T2 statistic implemented (posterior predictive check of covariance structure) o Posterior predictive checks with parallel computation using multiple cores o Bug fix for posterior predictive with fixed effects Changes in version 0.4.8 o New function posteriorPredictive() to draw samples of individual frequencies from posterior o T1 statistic now computed outside of JAGS (more stable, smaller mcmc object, additional argument M.T1 to specifiy the number of posterior samples used) Changes in version 0.4.7 o New function plotPriorPost() to compare prior vs. posterior densities o Default prior for traitMPT changed to xi="dunif(0,10)" for stability Changes in version 0.4.6 o Parameter-specific hyperpriors for betaMPT (alpha and beta) and traitMPT (mu and xi) Changes in version 0.4.5 o Special case of a single hierarchical parameter in traitMPT() [Wishart reduces to chi^2] o Allow fixed effect MPT parameters (e.g., a single guessing parameter for all participants). Specified in restrictions = list("g=FE") o Bayesian p-value in getGroupMeans() to test whether group mean differs from overall mean o Estimation of correlations of theta parameters in betaMPT (based on the MCMC samples; no explicit prior) Changes in version 0.4.4 o New function extendMPT() to get additional MCMC samples for fitted traitMPT and betaMPT o Bug fixes if parameter appear twice in model equation (e.g., u*u) Changes in version 0.4.3 o Use own summary function (functions in runjags are slow and unstable) o Estimation of DIC requries argument "dic=TRUE" or can be estimated afterwards by: fit$dic <- extract(fit$runjags, "dic") Changes in version 0.4.2 o Using the package runjags instead of R2jags (better functionality, e.g., provides progress bar during parallel sampling; max.time for autojags) o Making the function summarizeMPT() visible to allow users to recompute nice MPT summaries after changing the mcmc object (e.g., after the exclusion of MCMC samples) Changes in version 0.4.1 o new argument "T1group" to compute T1 statistic separately for a grouping factor (e.g., experimental condition; can be one of the predictors in traitMPT) o changed name of parameter estimate plotting function to "plotParam()" o new generic plotting function plot() for betaMPT and traitMPT (a convenient wrapper for the convergence plots in coda) Changes in version 0.4.0 o Possible to sample correlations AND predictors in traitMPT (using covStructure vs. predStructure) o Defaults for traitMPT: No predictors ; correlations for all covariates that are not included in predStructure o Changed argument name "covType" to "predType" (since it is only relevant for predStructure in traitMPT) o New argument corProbit to specify whether to compute correlations for probability- or probit-scaled MPT parameters o Allow to round to specific number of digits, e.g.: summary(fittedModel, round=6) o New function getParam() to conveniently extract posterior estimates (e.g., posterior mean, median, sd) o New function getGroupMeans() to get group estimates in traitMPT with discrete predictors (for single factors or combinations) o Updated vignette Changes in version 0.3.5 o Back to comma-separated data format for 'data.csv' and 'covData.csv' o Less informative Cauchy prior as default for continuous predictors in traitMPT o Additional argument IVprec in traitMPT to specify hyperprior for precision of continuous slope paramters o Data-generating function genMPT using general matrices of individual parameters o Parameter labels for output in fittedModel$mcmc$BUGSoutput Changes in version 0.3.4 o Support for WinBUGS and OpenBUGS removed o New function plotFreq to plot individual and mean raw frequencies per tree Changes in version 0.3.3 o autojags fixed o BCI and R^hat included in individual statistics o Better checks for input arguments for betaMPT, traitMPT Changes in version 0.3.2 o csv-format for "data" and "covData": semicolon (;) instead of comma (,) to separate cells o Printing of summary output to "parEstFile" improved Changes in version 0.3.1 o New function plotDistribution() to plot histograms of individual mean estimates Changes in version 0.3.0 o Discrete factors as covariates in traitMPT with fixed and random effects (see argument: covType) o Improved covariate handling: irrelevant columns are neglected o Goodness of fit plots for mean frequencies (plotFit) o Removed default values for arguments that were NULL previously o Updated vignette o Various Bugfixes Changes in version 0.2.3 o Predictors can be included in traitMPT (same arguments covData and covStructure as in betaMPT) o Nice summary for covariates in betaMPT and predictors in traitMPT o Checked and fixed data generation and fitting for latent trait and beta MPT model o Remaining issue: SD of parameters in betaMPT not precisely estimated o Informative error message if N=0 in a tree for a person o Updated vignette Changes in version 0.2.2 o Covariates can be included in betaMPT (see arguments covData and covStructure) o Data generation for latent trait MPT model: genTraitMPT() o Example model files (2HTM and 2HTSM) in library path: /TreeBUGS/MPTmodels/ o Examples for readEQN and data generation (genBetaMPT, genTraitMPT) Changes in version 0.2.0 o Implementation of Beta-MPT and latent-trait MPT model: betaMPT() ; traitMPT() o Sample and summarize transformed parameters (e.g., "deltaD=d1-d2") o Posterior predictive checks (T1 statistic for group and individual data) o Basic summary and plotting functionality o Functions to generate data according to the Beta-MPT: genBetaMPT() o Package vignette with examples: vignette("TreeBUGS") o Checking EQN file for consistency (identifiability etc.): readEQN()