simcdm
Simulate Cognitive Diagnostic Model ('CDM') Data
Provides efficient R and 'C++' routines to simulate cognitive diagnostic
model data for Deterministic Input, Noisy "And" Gate ('DINA') and
reduced Reparameterized Unified Model ('rRUM') from
Culpepper and Hudson (2017)
-brightgreen.svg?style=flat){kind=icon}
The goal of simcdm is to provide flexible ways to simulate data under
cognitive diagnostic models.
Installation
You can install simcdm from GitHub with:
# install.packages("devtools")devtools::install_github("tmsalab/simcdm")
Usage
To use simcdm, load the package using:
library("simcdm")
Overview
There are four distinct sets of functions within the package:
- Attributes:
attribute_classes(),attribute_bijection(),attribute_inv_bijection(), andsim_subject_attributes(). - Matrix:
sim_q_matrix()andsim_eta_matrix() - Deterministic Input, Noisy And Gate (DINA):
sim_dina_items()andsim_dina_attributes() - reduced Reparameterized Unified Model (rRUM):
sim_rrum_items()
Functions that use random numbers to simulate values are named with the
prefix of sim_*(). This is done to allow for functions to be quickly
identified and used through autocomplete inside of the RStudio
IDE. At a later time, the attribute_*() will
likely be moved to a different package.
For more details, please see the package vignettes:
Authors
James Joseph Balamuta and Steven Andrew Culpepper with contributions from Aaron Hudson.
Citing the simcdm package
To ensure future development of the package, please cite simcdm
package if used during the analysis or simulations. Citation information
for the package may be acquired by using in R:
citation("simcdm")
License
GPL (>= 2)
News
simcdm 0.1.1
Changes
- Addressed a signed vs. unsigned integer comparison.
Documentation
- Fixed spacing issues brought on from an organization of C++ code.
Deployment
- Changed unit tests to use R 3.5.0's RNG setup in anticipation for the RNG change in R 3.6.0.
simcdm 0.1.0
Changes
- Renamed
sim_attribute_classes()toattribute_classes(). - Addressed ambiguous calls to
std::pow(<int>, <int>)
Documentation
- Added a
CITATIONfile for the package - Improved Vignette Examples
- Improved README contents
Deployment
- Added Unit Tests to verify simulation routines and attribute generations.
- Added testing on Travis-CI for the previous release of R, e.g. the
oldrel.
simcdm 0.0.5
Features
- Added C++ and R functions for simulation of:
- Deterministic Input, Noisy "And" Gate (DINA)
- Item Response:
sim_dina_items() - $\eta$ Response:
sim_dina_attributes().
- Item Response:
- reduced Reparameterized Unified Model (rRUM):
- Item Reponse:
sim_rrum_items()
- Item Reponse:
- Matrices:
- Random Q Matrix:
sim_q_matrix() - ETA Matrix:
sim_eta_matrix() - Latent Attribute Profiles for Subjects:
sim_subject_attributes()
- Random Q Matrix:
- Attributes:
- Latent Attribute Profile Matrix:
sim_attribute_classes() - Attribute Bijection:
attribute_bijection() - Attribute Inverse Bijection:
attribute_inv_bijection()
- Latent Attribute Profile Matrix:
- Deterministic Input, Noisy "And" Gate (DINA)