The R Project SVN R

Rev

Rev 85829 | Show entire file | Ignore whitespace | Details | Blame | Last modification | View Log | RSS feed

Rev 85829 Rev 85835
Line 42... Line 42...
42 
\details{
 42 
\details{
43 
  By default, \code{method = "unbiased"},
 43 
  By default, \code{method = "unbiased"},
44 
  The covariance matrix is divided by one minus the sum of squares of
 44 
  The covariance matrix is divided by one minus the sum of squares of
45 
  the weights, so if the weights are the default (\eqn{1/n}) the conventional
 45 
  the weights, so if the weights are the default (\eqn{1/n}) the conventional
46 
  unbiased estimate of the covariance matrix with divisor \eqn{(n - 1)}
 46 
  unbiased estimate of the covariance matrix with divisor \eqn{(n - 1)}
47 
  is obtained.  This differs from the behaviour in S-PLUS which
 47 
  is obtained.
48 
  corresponded to \code{method = "ML"} and does not divide.
 - 
 
49 
}
 48 
}
50 
\seealso{\code{\link{cov}} and \code{\link{var}}.
 49 
\seealso{\code{\link{cov}} and \code{\link{var}}.
51 
}
 50 
}
52 
\examples{
 51 
\examples{
53 
 (xy <- cbind(x = 1:10, y = c(1:3, 8:5, 8:10)))
 52 
 (xy <- cbind(x = 1:10, y = c(1:3, 8:5, 8:10)))