\name{bdgraph.npn}
\alias{bdgraph.npn}

\title{Nonparametric transfer}

\description{ This function transfers non-Gaussian data to Gaussian.  }

\usage{ bdgraph.npn( data, npn = "shrinkage", npn.thresh = NULL ) }

\arguments{
  \item{data}{ It could be a (\eqn{n \times p}{n x p}) matrix or a data.frame corresponding to the data. }
  \item{npn}{ 
  A character with three options \code{"shrinkage"} (default), \code{"truncation"}, and \code{"skeptic"}. Option \code{"shrinkage"} is for the 
  shrunken transformation, option \code{"truncation"} is for the truncated transformation and option \code{"skeptic"} is for the 
  non-paranormal skeptic transformation. For more details see references.
  }
  \item{npn.thresh}{ The truncation threshold; it is only for the truncated transformation (when
  \code{npn = } \code{"truncation"}). The default value is \eqn{1/(4n^{1/4} \sqrt{\pi \log(n)})}. }
}

\value{
\item{data}{ A (\eqn{n \times p}) matrix of transferred data, if \code{npn = "shrinkage" or "truncation"}, and 
  a non-paranormal correlation (\eqn{p \times p}) matrix, if \code{npn = "skeptic"}.  
  }
}

\references{
Liu, H., F. Han, M. Yuan, J. Lafferty, and L. Wasserman (2012). High Dimensional Semiparametric Gaussian Copula Graphical Models, \emph{Annals of Statistics}  40(4):2293-2326

Zhao, T. and H. Liu (2012). The \pkg{huge} Package for High-dimensional Undirected Graph Estimation in \code{R}, \emph{Journal of Machine Learning Research}, 13:1059-1062
}

\author{ Abdolreza Mohammadi and Ernst Wit \cr }

\seealso{ \code{\link{bdgraph.sim}} and \code{\link{bdgraph}} }

\examples{
\dontrun{
# generating multivariate normal data from a 'random' graph
data.sim <- bdgraph.sim( n = 6, p = 4, size = 4 )
data     <- ( data.sim $ data - 3 ) ^ 4
data

# transfer the data by truncation 
bdgraph.npn( data, npn = "truncation" )

# transfer the data by shrunken 
bdgraph.npn( data, npn = "shrunken" )

# transfer the data by skeptic 
bdgraph.npn( data, npn = "skeptic" )
}
}