Hierarchical Bayesian ANOVA Models

It covers several Bayesian Analysis of Variance (BANOVA) models used in analysis of experimental designs in which both within- and between- subjects factors are manipulated. They can be applied to data that are common in the behavioral and social sciences. The package includes: Hierarchical Bayes ANOVA models with normal response, t response, Binomial (Bernoulli) response, Poisson response, ordered multinomial response and multinomial response variables. All models accommodate unobserved heterogeneity by including a normal distribution of the parameters across individuals. Outputs of the package include tables of sums of squares, effect sizes and p-values, and tables of predictions, which are easily interpretable for behavioral and social researchers. The floodlight analysis and mediation analysis based on these models are also provided. BANOVA uses 'Stan' and 'JAGS' as the computational platform.


News

BANOVA v1.1.1 (Release date: 2018-01-21)

  • Added vignettes

BANOVA v1.1.0 (Release date: 2018-01-04)

  • Added R2 measure of fit
  • Added level-1 variance term to the calculation of R2 and effect sizes
  • Added p values and effect sizes for indirect effects
  • BANOVA mediation only prints indirect effects, users can still access direct effects through the dir_effects element
  • BANOVA mediation for Multinomial models is under development
  • Use 3 decimals for the 95%CI of the effect sizes
  • Default settings for MCMC burn-in=5000, sample=2000, thin=10
  • Other variables in the model are included in the output of floodlight analysis
  • Users can set up values for other numeric variables for the floodlight analysis

BANOVA v1.0.0 (Release date: 2017-11-02)

  • add seeds for replications

BANOVA v0.9.1 (Release date: 2017-08-26)

  • single level models are supported for old models except Binomial and Bernoulli
  • fix the bug that the interaction between the mediator and moderators is missing in the mediation output
  • The names in the table of indirect effets are reordered (alphabetically)

BANOVA v0.9 (Release date: 2017-06-27)

  • BANOVA uses Stan!
  • all two-level models are moved into a simple function call: BANOVA.run
  • old two-level models (e.g. BANOVA.Normal, BANOVA.Bin, etc.) are still kept, but will be removed gradually
  • single level models are included

BANOVA v0.8 (Release date: 2016-12-12)

  • change the name 'table of means' to 'table of predictions'
  • column names of the data goalstudy are changed to make more sense
  • add floodlight analysis
  • add mediation analysis (single mediator, multiple moderators)

BANOVA v0.7 (Release date: 2016-08-24)

  • Exclude numeric variables for predictions
  • Change the format of convergence diag.
  • Change the calculations in the table of means and predictions to exp(mu + sigma^2/2) instead of exp(mu) for the Poisson model
  • Table of means is now included in the summary
  • Use 4 digits for the outputs in the summary except the table of means
  • Median is changed to mean in the output of table of means
  • Predictors that have 3 levels or fewer will be automatically considered as factors

BANOVA v0.6 (Release date: 2016-08-01)

  • Check missing data for independent variables
  • Fix typos
  • Add warnings for mean centering of numeric variables
  • Print full convergence diag. for the Heidelberg and Welch diagnostic.
  • Table of means can now print any level of interactions

Reference manual

It appears you don't have a PDF plugin for this browser. You can click here to download the reference manual.

install.packages("BANOVA")

1.1.2 by Chen Dong, 22 days ago


Browse source code at https://github.com/cran/BANOVA


Authors: Chen Dong , Michel Wedel


Documentation:   PDF Manual  


GPL (>= 2) license


Imports rjags, runjags, coda, rstan, methods

Suggests knitr, rmarkdown

System requirements: C++11, JAGS >= 4.1.0 (http://mcmc-jags.sourceforge.net)


See at CRAN