modelvec: Univariate and Multivariate Model Names

In upclass: Updated Classification Methods using Unlabeled Data

Description

Model names to be used in the upclass package for univariate and multivariate data.

Usage

modelvec(d = 1)

Arguments

d	The dimension of the data. By default, d=1, and the data is considered to be univariate.

Value

if d=1, returned is a vector with the first two of the following components only; otherwise, they are omitted and the vector contains the remaining components:

"E"	Univariate, equal variance
"V"	Univariate, variable variance
"EII"	Multivariate, equal volume and spherical
"VII"	Multivariate, variable volume and spherical
"EEI"	Multivariate, equal volume, equal shape and axis aligned
"VEI"	Multivariate, variable volume, equal shape and axis aligned
"EVI"	Multivariate, equal volume, variable shape and axis aligned
"VVI"	Multivariate, variable volume, variable shape and axis aligned
"EEE"	Multivariate, equal volume, equal shape and equal orientation
"EEV"	Multivariate, equal volume, equal shape and variable orientation
"VEV"	Multivariate, variable volume, equal shape and variable orientation
"VVV"	Multivariate, variable volume, variable shape and variable orientation

References

Banfield, J.D. and Raftery, A.E. (1993). Model based Gaussian and non-gaussian clustering. Biometrics, 49 (3): 803-821.

Fraley, C. and Raftery, A.E. (2002). Model-based clustering, discriminant analysis, and density estimation. Journal of the Americal Statistical Association 97 (458), 611-631.

See Also

upclassify, noupclassify.

Examples

modelvec(1) # Models available for univariate data.

data(iris)
modelvec(ncol(iris[,-5])) # Models available for multivariate data

upclass documentation built on May 29, 2017, 5:12 p.m.