plot2d.cv.scca.Rd

% Generated by roxygen2: do not edit by hand
% Please edit documentation in R/scca.R
\name{plot2d.cv.scca}
\alias{plot2d.cv.scca}
\title{Plot the results of SCCA cross-validation (Pearson correlation) as a 2D
surface. The optimal position is shown as a cross ('+').}
\usage{
\method{plot2d}{cv.scca}(x, dim = 1, plot = FALSE, ...)
}
\arguments{
\item{x}{An object of class "cv.scca"}

\item{dim}{Integer. Which dimension to plot (all will be plotted by
default).}

\item{plot}{Logical. Whether to plot (print) the ggplot2 object or just
return it.}
}
\value{
the ggplot2 object
}
\description{
Plot the results of SCCA cross-validation (Pearson correlation) as a 2D
surface. The optimal position is shown as a cross ('+').
}
\details{
Plots the cross-validated Pearson correlation, as a 2D surface with the
   lambda1 and lambda2 penalties on the axes.
}
\examples{

#######################
## HapMap3 chr1 example
data(hm3.chr1)
X <- scale2(hm3.chr1$bed)
n <- nrow(X)
m <- ncol(X)
k <- 5
B <- matrix(stats::rnorm(m * k), m, k)
Y <- X \%*\% B + stats::rnorm(n * k)

r <- cv.scca(X, Y,
   standx = "sd", standy = "sd", nfolds = 3, ndim = 2,
   lambda1 = seq(1e-3, 1e-1, length = 10),
   lambda2 = seq(1e-4, 0.5, length = 8)
)

g <- plot2d(r, dim = 1, plot = FALSE)
print(g)


}