\name{drawInitialLatent}
\alias{drawInitialLatent}
\alias{lpriorLatent}
\title{Draws initial values for a latent variable or calculates its log prior
}
\description{

  The function \code{drawIntialLatent} takes a proficiency model and an
  associated parameter object and generates the specified number of
  random latent profiles.

  The function \code{lpriorLatent} takes a proficiency model, its
  associated parameter, and a matrix of latent profiles and calculates
  the log priors for those profiles.
  
}
\usage{
drawInitialLatent(pm, ncases = 1, background = NA, param=parameters(pm))
lpriorLatent(pm, latent, background = NA, param=parameters(pm))
}
\arguments{
  \item{pm}{
    An object of class \code{\link{ProficiencyModel}}.
  }
  \item{ncases}{
    The number of latent profiles to be generated.
  }
  \item{latent}{
    A matrix of latent vectors whose likelihood is to be evaluated.
    It should be true that \code{ncol(obs) == nlatent(pm)}.
  }
  \item{background}{
    An optional set of background (demographic) variables.  Some
    proficiency model types ignore background variables.
  }
  \item{param}{
    An object representing the parameters of \code{pm}.  The type
    required is determined by the class of \code{pm}.
  }
}
\details{

  The function \code{drawIntialLatent} draws the initial latent values.
  It outputs a matrix whose rows are cases, and whose columns represent
  the latent variables.

  The function \code{lpriorLatent} evaluates the prior distribution at
  each row of the \code{latent} matrix.  A log prior value is generated
  for each row of \code{latent}.

  Each concrete subclass of \code{ProficiencyModel} has a corresponding
  parameter class.  The class of the \code{param} argument should be the
  appropriate parameter class for the \code{pm} argument.
}
\value{

  The function \code{drawIntialLatent} returns a matrix with
  \code{ncases} rows and \code{nlatent(pm)} columns.

  The function \code{lpriorLatent} returns a vector log prior values
  with a number of elements equal to \code{nrow(latent)}.
  
}
\references{
%% ~put references to the literature/web site here ~
}
\author{Russell Almond}
\seealso{
  \code{\link{ProficiencyModel}}
  
}
\examples{
pm1 <-
  TimelessNormalPM(muMean=c(Mechanics=2,Fluency=2), varWeight=3,
                   precScale=solve(matrix(c(.7,.3,.3,.7),2,2)), Sdf=3)

parameters(pm1) <- drawPMParam(pm1)

pm1.stud10 <- drawInitialLatent(pm1,10)
lpriorLatent(pm1,pm1.stud10)

}
\keyword{ distribution }
\keyword{ random }