dens.prod.ordi {LCAextend}

computes the probability of a given discrete measurement vector for all classes under a product of multinomial

Description

computes the probability of an individual's discrete measurement vector for all latent classes under a multinomial distribution product, eventually taking covariates into account. This is an internal function not meant to be called by the user.

Usage

dens.prod.ordi(y.x, param, var.list = NULL)

Arguments

y.x

a vector y of values of the ordinal variables (measurements) followed by the values x of covariates, if any,

param

a list of the parameters alpha (cumulative logistic coefficients), see init.ordi,

var.list

a list of integers indicating which covariates (taken from x) are used for a given type of measurement.

Details

If there are K latent classes, d measurements and each measurement has S[j] possible values, alpha is a list of d elements, each is a K times S[j]+length{var.list[[j]]} matrix. For a class C=k, dens[k]=\displaystyle\prod_{j=1}^d{P(Y_j=y_j|C=k,X_j=x_j)}, where P(Y_j=y_j|C=k,X_j=x_j) is computed from the cumulative logistic coefficients alpha[[j]][k,] and covariates x[var.list[[j]]],

Value

The function returns a vector dens of length K, where dens[k] is the probability of measurement vector y with covariates x, if the individual belongs to class k.

See Also

See Also init.ordi,

Examples

#data
data(ped.ordi)
status <- ped.ordi[,6]
y <- ped.ordi[status==2,7:ncol(ped.ordi)]
#param
data(param.ordi)
#the function applied for measurement of the first individual in the ped.ordi
dens.prod.ordi(y.x=y[1,],param.ordi)