Bayesian Estimation of the Reduced Reparameterized Unified Model with Gibbs Sampling

Implementation of Gibbs sampling algorithm for Bayesian Estimation of the Reduced Reparameterized Unified Model ('rrum'), described by Culpepper and Hudson (2017) .


BuildStatus Package-License CRAN VersionBadge CRANStatus RStudio CRAN Mirror’s MonthlyDownloads RStudio CRAN Mirror’s TotalDownloads Coveragestatus

The goal of rrum is to provide an implementation of Gibbs sampling algorithm for Bayesian Estimation of Reduced Reparameterized Unified Model (rrum), described by Culpepper and Hudson (2017) <doi: 10.1177/0146621617707511>.

Installation

You can install rrum from CRAN using:

install.packages("rrum")

Or, you can be on the cutting-edge development version on GitHub using:

if(!requireNamespace("devtools")) install.packages("devtools")
devtools::install_github("tmsalab/rrum")

Usage

To use rrum, load the package using:

library("rrum")
#> Loading required package: simcdm

From here, the rRUM model can be estimated using:

rrum_model = rrum(<data>, <q>)

Additional parameters can be accessed with:

rrum_model = rrum(<data>, <q>, chain_length = 10000L,
                  as = 1, bs = 1, ag = 1, bg = 1,
                  delta0 = rep(1, 2^ncol(Q)))

rRUM item data can be simulated using:

# Set a seed for reproducibility
set.seed(888)
 
# Setup Parameters
N = 15   # Number of Examinees / Subjects
J = 10   # Number of Items
K = 2    # Number of Skills / Attributes
 
# Simulate identifiable Q matrix
Q = sim_q_matrix(J, K)
 
# Penalties for failing to have each of the required attributes
rstar  = .5 * Q
 
# The probabilities of answering each item correctly for individuals 
# who do not lack any required attribute
pistar = rep(.9, J)
 
# Latent Class Probabilities
pis = c(.1, .2, .3, .4)
 
# Generate latent attribute profile with custom probability (N subjects by K skills)
subject_alphas = sim_subject_attributes(N, K, prob = pis)
 
# Simulate rrum items
rrum_items = simcdm::sim_rrum_items(Q, rstar, pistar, subject_alphas)

Authors

Steven Andrew Culpepper, Aaron Hudson, and James Joseph Balamuta

Citing the rrum package

To ensure future development of the package, please cite rrum package if used during an analysis or simulation study. Citation information for the package may be acquired by using in R:

citation("rrum")

License

GPL (>= 2)

News

rrum 0.2.0

API Breakage

  • Deprecated rRUM_Gibbs() in favor of rrum().
  • Deprecated pi_reference() in favor of simcdm::attribute_classes().

Changes

  • Added CITATION file for citing both the APM paper and package.
  • Imported simulation functions from simcdm

Documentation

  • Improved README examples

Deployment

  • Added Travis-CI configuration for TMSA Lab.
  • Added Unit Tests for model reproducibility.
  • Added code coverage checks.

rrum 0.1.0

  • Improved documentation
  • Addressed RcppExport updates

rrum 0.0.5

  • Introduced new rRUM estimation routine.
  • Provided a means to simulate rRUM data.

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("rrum")

0.2.0 by James Joseph Balamuta, 7 months ago


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


Authors: Steven Andrew Culpepper [aut, cph] , Aaron Hudson [aut, cph] , James Joseph Balamuta [aut, cph, cre]


Documentation:   PDF Manual  


Task views: Psychometric Models and Methods


GPL (>= 2) license


Imports Rcpp

Depends on simcdm

Suggests testthat, covr

Linking to Rcpp, RcppArmadillo, rgen, simcdm

System requirements: C++11


Enhanced by CDM.


See at CRAN