The R Project SVN R

\name{solve}
\title{Solve a System of Equations}
\usage{
solve(a, b, \dots)

\method{solve}{default}(a, b, tol, LINPACK = FALSE, \dots)
}
\alias{solve}
\alias{solve.default}
\arguments{
  \item{a}{a square numeric or complex matrix containing the coefficients of
    the linear system.}
  \item{b}{a numeric or complex vector or matrix giving the right-hand
    side(s) of the linear system.  If missing, \code{b} is taken to be
    an identity matrix and \code{solve} will return the inverse of \code{a}.}
  \item{tol}{the tolerance for detecting linear dependencies in the
    columns of \code{a}.  If \code{LINPACK} is \code{TRUE} the default
    is \code{1e-7}, otherwise it is \code{.Machine$double.eps}. Future
    versions of R may use a tighter tolerance. Not presently used with
    complex matrices \code{a}.}
  \item{LINPACK}{logical. Should LINPACK be used (for compatibility with
    \R < 1.7.0)?}
  \item{\dots}{further arguments passed to or from other methods}
}
\description{
  This generic function solves the equation \code{a \%*\% x = b} for \code{x},
  where \code{b} can be either a vector or a matrix.
}
\details{
  As from \R 1.3.0, \code{a} or \code{b} can be complex, in
  which case LAPACK routine \code{ZESV} is used.
  This uses double complex arithmetic which might not be available on all
  platforms.

  The row and column names of the result are taken from the column
  names of \code{a} and of \code{b} respectively.  As from \R 1.7.0
  if \code{b} is missing the column names of the result are the row
  names of \code{a}.  No check is made that the column names of \code{a}
  and the row names of \code{b} are equal.

  For back-compatibility \code{a} can be a (real) QR decomposition,
  although \code{\link{qr.solve}} should be called in that case.
  \code{\link{qr.solve}} can handle non-square systems.
}
\references{
  Becker, R. A., Chambers, J. M. and Wilks, A. R. (1988)
  \emph{The New S Language}.
  Wadsworth \& Brooks/Cole.
}
\seealso{
  \code{\link{solve.qr}} for the \code{qr} method,
  \code{\link{backsolve}}, \code{\link{qr.solve}}.
}
\examples{
hilbert <- function(n) { i <- 1:n; 1 / outer(i - 1, i, "+") }
h8 <- hilbert(8); h8
sh8 <- solve(h8)
round(sh8 \%*\% h8, 3)

A <- hilbert(4)
A[] <- as.complex(A)
## might not be supported on all platforms
try(solve(A))
}
\keyword{algebra}