The R Project SVN R

\name{stl}
\alias{stl}
\alias{print.stl}
\alias{plot.stl}
\title{Seasonal Decomposition of Time Series by Loess}
\usage{
stl(x, s.window = NULL, s.degree = 0, t.window = NULL, t.degree = 1,
    robust = FALSE, na.action = na.fail)
}
\arguments{
  \item{x}{A univariate time series to be decomposed.
    This should be an object of class \code{"ts"} with a frequency
    greater than one.}
  \item{s.window}{Either the string \code{"periodic"} or the span (in
    lags) of the loess window for seasonal extraction, which should
    be odd.  This has no default.}
  \item{s.degree}{Degree of locally-fitted polynomial in seasonal
    extraction.  Should be zero or one.} 
  \item{t.window}{The span (in lags) of the loess window for trend
    extraction, which should be odd.  There is a reasonable
    default.}
  \item{t.degree}{Degree of locally-fitted polynomial in trend
    extraction.  Should be zero or one.}
  \item{robust}{Should robust fitting be used in the \code{loess}
    procedure?}
  \item{na.action}{Action on missing values.}
}
\description{
  Decompose a time series into seasonal, trend and irregular
  components.
}
\details{
  The seasonal component is found by \emph{loess} smoothing the
  seasonal sub-series (the series of all January values, \ldots); if
  \code{s.window = "periodic"} smoothing is effectively replaced by
  taking the mean. The seasonal values are removed, and the remainder
  smoothed to find the trend. The overall level is removed from the
  seasonal component and added to the trend component. This process is
  iterated a few times.  The \code{remainder} component is the
  residuals from the seasonal plus trend fit.
}
\value{
  An object of class \code{"stl"} with components
  \item{time.series}{a multiple time series with columns
    \code{seasonal}, \code{trend} and \code{remainder},}
  \item{weights}{the final robust weights (all one if fitting is not
    done robustly,}
  \item{call}{the matched call.}
}
\references{
  R. B. Cleveland, W. S. Cleveland, J.E.  McRae, and I. Terpenning
  (1990).
  STL:  A  Seasonal-Trend  Decomposition  Procedure Based on Loess.
  \emph{Journal of Official Statistics}, \bold{6}, 3--73.
}
\author{B.D. Ripley; Fortran code by Cleveland \emph{et al.} (1990) from
  \file{netlib}.}

\note{This is similar to but not identical to the \code{stl} function in
  S-PLUS. The \code{remainder} component given by S-PLUS is the sum of
  the \code{trend} and \code{remainder} series from this function.}

\seealso{\code{\link{loess}} in package \file{modreg} (which is not
  actually used in \code{stl}).}

\examples{
data(nottem)
plot(stl(nottem, "per"))
data(co2)
plot(stl(log(co2), s.window=21))
## linear trend, strict period.
plot(stl(log(co2), s.window="per", t.window=1000))
}
\keyword{ts}