stuart-package: STUART: Subtests Using Algorithmic Rummaging Techniques
In stuart: Subtests Using Algorithmic Rummaging Techniques
stuart-package R Documentation
STUART: Subtests Using Algorithmic Rummaging Techniques
Description
The STUART-Package automates the generation of subtests from
a given set of items within the confines of confirmatory factor analysis.
Functionality
Using this package subtests can be generated in four different ways: using a pseudo-random approach rooted in Ant-Colony-Optimization via the mmas-function, using a simple genetic algorithm via the gene-function, using a brute-force approach via the aptly named bruteforce-function, or by random chance, using the randomsamples-function.
Addtionally, there are some convenience functions which are more or less useful. The combinations-function can be used to determine the number of possible subtests to inform a decision on which selection approach to use. The crossvalidate-function can be used to evaluate the quality of a selection in a different (sub-)sample. To add to this functionality, the holdout-function randomly splits the data into a calibration and a validation sample. The entire process can be applied to k samples with kfold. The heuristics-function can be used to extract the formatting of heuristic matrices which can be provided to the mmas-function.
As of version 0.10.0 this package also includes some convenience functions for handling objective functions. These are mainly fixedobjective - to generate a fixed objective function containing any number of a variety of possible quality criteria - and empiricalobjective - to generate an adaptive objective function based on the quality determined for previous solution on any such criteria. In addition objectivematrices provides functionality to extract matrices for situations in which, for example, latent correlations or regressions predicting distal outcomes are included into the selection procedure.
The package also provides three datasets to try things out with: fairplayer, sia, and sups.
Author(s)
Maintainer: Martin Schultze schultze@psych.uni-frankfurt.de
Other contributors:
Johanna Schüller [contributor]
stuart documentation built on June 7, 2023, 6:12 p.m.
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
| stuart-package | R Documentation |
STUART: Subtests Using Algorithmic Rummaging Techniques
Description
The STUART-Package automates the generation of subtests from a given set of items within the confines of confirmatory factor analysis.
Functionality
Using this package subtests can be generated in four different ways: using a pseudo-random approach rooted in Ant-Colony-Optimization via the mmas-function, using a simple genetic algorithm via the gene-function, using a brute-force approach via the aptly named bruteforce-function, or by random chance, using the randomsamples-function.
Addtionally, there are some convenience functions which are more or less useful. The combinations-function can be used to determine the number of possible subtests to inform a decision on which selection approach to use. The crossvalidate-function can be used to evaluate the quality of a selection in a different (sub-)sample. To add to this functionality, the holdout-function randomly splits the data into a calibration and a validation sample. The entire process can be applied to k samples with kfold. The heuristics-function can be used to extract the formatting of heuristic matrices which can be provided to the mmas-function.
As of version 0.10.0 this package also includes some convenience functions for handling objective functions. These are mainly fixedobjective - to generate a fixed objective function containing any number of a variety of possible quality criteria - and empiricalobjective - to generate an adaptive objective function based on the quality determined for previous solution on any such criteria. In addition objectivematrices provides functionality to extract matrices for situations in which, for example, latent correlations or regressions predicting distal outcomes are included into the selection procedure.
The package also provides three datasets to try things out with: fairplayer, sia, and sups.
Author(s)
Maintainer: Martin Schultze schultze@psych.uni-frankfurt.de
Other contributors:
Johanna Schüller [contributor]
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
Embedding an R snippet on your website
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.