poLCA.predcell {poLCA} | R Documentation |
Calculates the predicted cell percentages from a latent class model, for specified values of the manifest variables.
poLCA.predcell(lc,y)
lc |
A model object estimated using the |
y |
A vector or matrix containing series of responses on the manifest variables in |
The parameters estimated by a latent class model can be used to produce a density estimate of the underlying probability mass function across the cells in the multi-way table of manifest variables. This function calculates cell percentages for that density estimate, corresponding to selected sets of responses on the manifest variables, y
.
A vector containing cell percentages corresponding to the specified sets of responses y
, based on the estimated latent class model lc
.
data(carcinoma)
f <- cbind(A,B,C,D,E,F,G)~1
lca3 <- poLCA(f,carcinoma,nclass=3) # log-likelihood: -293.705
# Only 20 out of 32 possible response patterns are observed
lca3$predcell
# Produce cell probabilities for one sequence of responses
poLCA.predcell(lc=lca3,y=c(1,1,1,1,1,1,1))
# Estimated probabilities for a cell with zero observations
poLCA.predcell(lc=lca3,y=c(1,1,1,1,1,1,2))
# Cell probabilities for both cells at once; y entered as a matrix
poLCA.predcell(lc=lca3,y=rbind(c(1,1,1,1,1,1,1),c(1,1,1,1,1,1,2)))