predictg: Cluster Prediction

In ContaminatedMixt: Clustering and Classification with the Contaminated Normal

Cluster Prediction

Description

Cluster prediction for multivariate observations based on uncontaminated/contaminated normal mixture models

Usage

CNpredict(newdata, prior, mu, invSigma, eta=NULL, alpha=NULL)
## S3 method for class 'ContaminatedMixt'
predict(object, newdata, ...)

Arguments

newdata	a dim=c(n,p) matrix representing the coordinates of n new data point(s)
object	an object of class ContaminatedMixt resulting from a call to CNmixt. When several models have been estimated, getBestModel is used to select one of them
...	Options to be passed to getBestModel
prior	a vector with length=G, where G is the number of components of the mixture model. Its kth component is the mixing proportion for the kth component
mu	a dim=c(p,G) matrix with mean values for each component of the mixture model
invSigma	an array with dim=c(p,p,G) whose element invSigma[,,k] is the inverse covariance matrix for the kth component of the mixture model.
alpha	a vector of length=G with the proportions of good observations; it must be a number between 0 and 1. Use NULL for uncontaminated models
eta	a vector of length=G with the degree of contamination; it should be a number greater than 1. Use NULL for uncontaminated models

Value

a vector with group membership

Author(s)

Antonio Punzo, Angelo Mazza, Paul D. McNicholas

References

Punzo A., Mazza A. and McNicholas P. D. (2018). ContaminatedMixt: An R Package for Fitting Parsimonious Mixtures of Multivariate Contaminated Normal Distributions. Journal of Statistical Software, 85(10), 1–25.

Punzo A. and McNicholas P. D. (2016). Parsimonious mixtures of multivariate contaminated normal distributions. Biometrical Journal, 58(6), 1506–1537.

See Also

ContaminatedMixt-package

Examples

point <- c(0,0,0)
mu <- c(1,-2,3)
Sigma <- diag(3)
alpha <- 0.8
eta <- 5
f <- dCN(point, mu, Sigma, alpha, eta)
x <- rCN(10, mu, Sigma, alpha, eta)

ContaminatedMixt documentation built on May 31, 2023, 6:44 p.m.