| Line 57... |
Line 57... |
| 57 |
|
57 |
|
| 58 |
\item{weights}{numeric vector of non-negative observation weights,
|
58 |
\item{weights}{numeric vector of non-negative observation weights,
|
| 59 |
hence of same length as \code{x}. The default \code{NULL} is
|
59 |
hence of same length as \code{x}. The default \code{NULL} is
|
| 60 |
equivalent to \code{weights = rep(1/nx, nx)} where \code{nx} is the
|
60 |
equivalent to \code{weights = rep(1/nx, nx)} where \code{nx} is the
|
| 61 |
length of (the finite entries of) \code{x[]}. If \code{na.rm = TRUE}
|
61 |
length of (the finite entries of) \code{x[]}. If \code{na.rm = TRUE}
|
| 62 |
and there are \code{NA}'s in \code{x}, they \emph{and} the
|
62 |
and there are \code{NA}s in \code{x}, they \emph{and} the
|
| 63 |
corresponding weights are removed before computations. In that case,
|
63 |
corresponding weights are removed before computations. In that case,
|
| 64 |
when the original weights have summed to one, they are re-scaled to
|
64 |
when the original weights have summed to one, they are re-scaled to
|
| 65 |
keep doing so.
|
65 |
keep doing so.
|
| 66 |
|
66 |
|
| 67 |
Note that weights are \emph{not} taken into account for automatic
|
67 |
Note that weights are \emph{not} taken into account for automatic
|