initialize.para {rrMixture}

Initialization of Parameter Estimates

Description

‘initialize.para’ is used to initialize parameter estimates.

Usage

initialize.para(K, X, Y, ind0 = NULL,
                seed = NULL, km.nstart = 20, kmscale = FALSE, n.init = 100,
                commonvar = FALSE)

Arguments

K

number of mixture components.

X

n by p design matrix where n is the number of observations and p is the number of predictors.

Y

n by q response matrix where n is the number of observations and q is the number of responses.

ind0

vector of length n, specifying the initial assignment of the mixture membership of n observations when there is prior information on the membership. If ‘NULL’, K-means clustering technique is used to assign the membership for n observations. Default is ‘NULL’.

seed

seed number for the reproducibility of results. Default is ‘NULL’.

km.nstart

number of random sets considered to perform K-means clustering. Only used for K-means clustering. Default is 20.

kmscale

logical value, indicating whether Y is scaled prior to K-means clustering. Only used for K-means clustering. Default is ‘FALSE’.

n.init

number of initializations to try. Two methods for initial clustering are used: K-means and random clustering.

commonvar

logical value, indicating the homogeneity assumption of variance-covariance matrices across K mixture components. Default is ‘FALSE’.

Value

para

array of length K. It consists of K lists, each of which contains initial estimates of membership probability, coefficient matrix, and variance- covariance matrix.

Author(s)

Suyeon Kang, University of California, Riverside, [e-mail]; Weixin Yao, University of California, Riverside, [e-mail]; Kun Chen, University of Connecticut, [e-mail].

References

Kang, S., Chen, K., and Yao, W. (2022+). "Reduced rank estimation in mixtures of multivariate linear regression".

See Also

rrmix.sim.norm

Examples

#-----------------------------------------------------------#
# Simulation 1: Two Components Case
#-----------------------------------------------------------#
K2mod <- rrmix.sim.norm(K = 2, n = 100, p = 5, q = 5, rho = .5,
         b = 1, shift = 1, r.star = c(1, 3), sigma = c(1, 1),
         pr = c(.5, .5), seed = 1215)
K2ini <- initialize.para(K = 2, X = K2mod$X, Y = K2mod$Y,
         seed = 100)

#-----------------------------------------------------------#
# Simulation 2: Four Components Case
#-----------------------------------------------------------#

K4mod <- rrmix.sim.norm(K = 4, n = 600, p = 15, q = 15,
         rho = .5, b = 1, shift = 1, r.star = c(1, 1, 3, 3),
         sigma = c(1, 1, 1, 1), pr = c(.25, .25, .25, .25),
         seed = 1215)
K4ini <- initialize.para(K = 4, X = K4mod$X, Y = K4mod$Y,
         seed = 100)