init.ordi: computes the initial values for EM algorithm in the case of...

In LCAextend: Latent Class Analysis (LCA) with Familial Dependence in Extended Pedigrees

Description

computes the initial values of cumulative logistic coefficients alpha for the EM algorithm in the case of ordinal measurements and a product multinomial model.

Usage

init.ordi(y, K, x = NULL, var.list = NULL)

Arguments

y	a n times d matrix of ordinal (or discrete) measurements, where n is the number of individuals and d is the number of measurements. All entries must be finite, if not an error is produced,
K	number of latent classes of the model,
x	a matrix of covariates if any, default is NULL (no covariates),
var.list	list of integers indicating which covariates (taken from x) are used for a given measurement (a column of y).

Details

The function allocates every individual to a class and evaluates the cumulative logistic coefficients for each measurement and each class. Regression coefficients for the covariates are set to 0.

Value

The function returns a list of one element alpha which is a list of d elements, each element alpha[[j]] is a K times S-1 matrix, where S is the number of values of the measurement y[,j], a row alpha[[j]][k,] gives the the cumulative logistic coefficients of class k and measurement j using alpha.compute.

See Also

alpha.compute

Examples

#data
data(ped.ordi)
status <- ped.ordi[,6]
y <- ped.ordi[,7:ncol(ped.ordi)]
#the function
init.ordi(y[status==2,],K=3)

LCAextend documentation built on May 2, 2019, 2:02 a.m.