NominalVariables
is an option for statistical functions such as LinearModelFit that specifies which variables should be treated as having discrete values specified by names.
Details
- Possible settings for NominalVariables include:
-
Automatic automaticaly detect nominal variables All treat all variables as nominal None treat no variables as nominal {v1,v2,…} treat the variables vi as nominal - When NominalVariables->Automatic, different methods are used to detect nominal variables. In functions returning FittedModel[…], variables with non-numeric data values are treated as nominal.
- In linear models, a nominal variable v with possible values s1, s2, … is represented as a collection of variables with values 1 or 0 corresponding to each case v==si.
Examples
open allclose allBasic Examples (1)
Scope (2)
Automatically select nominal variables:
Specify nominal variables in GeneralizedLinearModelFit:
With LogitModelFit:
With ProbitModelFit:
Applications (3)
Compute the model estimates for the three groups:
Fit a two‐way ANOVA model with interaction term:
Compute estimates for each cell:
Visualize curves for the three x categories and label using Tooltip:
Text
Wolfram Research (2008), NominalVariables, Wolfram Language function, https://reference.wolfram.com/language/ref/NominalVariables.html (updated 2014).
CMS
Wolfram Language. 2008. "NominalVariables." Wolfram Language & System Documentation Center. Wolfram Research. Last Modified 2014. https://reference.wolfram.com/language/ref/NominalVariables.html.
APA
Wolfram Language. (2008). NominalVariables. Wolfram Language & System Documentation Center. Retrieved from https://reference.wolfram.com/language/ref/NominalVariables.html