vipCriterion: The VIP criterion
In larssnip/mpda: Classification in a multivariate setting
vipCriterion R Documentation
The VIP criterion
Description
Computes the VIP criterion used to rank variable importance.
Usage
vipCriterion(pls.model, dim = 1)
Arguments
pls.model
Object with fitted PLS-model.
dim
Integer, the number of dimensions to consider.
Details
After the fitting of a PLS-model, some of the original variables will have more
impact than the others on the prediction of the response. The VIP criterion is one way
to quantify this, see Chong&Yun, 2005. This criterion requires a single response regression
problem, which means a two-class classification problem.
A large VIP indicates the corresponding variable is important. A threshold at 1.0 is often
used, variables with VIP above 1.0 are the important ones.
Value
A vector of VIP scores, one for each variable in the predictor matrix of the fitted
PLS model.
Author(s)
Lars Snipen.
References
Chong, Il-Gyo & Jun, Chi-Hyuck, 2005, Performance of some variable selection
methods when multicollinearity is present, Chemometrics and Intelligent Laboratory
Systems 78, 103–112.
See Also
eliminator.
larssnip/mpda documentation built on March 28, 2022, 3:37 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.
| vipCriterion | R Documentation |
The VIP criterion
Description
Computes the VIP criterion used to rank variable importance.
Usage
vipCriterion(pls.model, dim = 1)
Arguments
pls.model |
Object with fitted PLS-model. |
dim |
Integer, the number of dimensions to consider. |
Details
After the fitting of a PLS-model, some of the original variables will have more impact than the others on the prediction of the response. The VIP criterion is one way to quantify this, see Chong&Yun, 2005. This criterion requires a single response regression problem, which means a two-class classification problem.
A large VIP indicates the corresponding variable is important. A threshold at 1.0 is often used, variables with VIP above 1.0 are the important ones.
Value
A vector of VIP scores, one for each variable in the predictor matrix of the fitted PLS model.
Author(s)
Lars Snipen.
References
Chong, Il-Gyo & Jun, Chi-Hyuck, 2005, Performance of some variable selection methods when multicollinearity is present, Chemometrics and Intelligent Laboratory Systems 78, 103–112.
See Also
eliminator.
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.