1\name{sm.ts.pdf}
2\alias{sm.ts.pdf}
3\title{
4Nonparametric density estimation of stationary time series data
5}
6\description{
7This function estimates the density function of a time series \code{x},
8assumed to be stationary. The univariate marginal density is estimated
9in all cases; bivariate densities of pairs of lagged values are estimated
10depending on the parameter \code{lags}.
11}
12\usage{
13sm.ts.pdf(x, h = hnorm(x), lags, maxlag = 1, ask = TRUE)
14}
15\arguments{
16\item{x}{
17a vector containing a time series
18}
19\item{h}{
20bandwidth
21}
22\item{lags}{
23  for each value, \code{k} say, in the vector \code{lags} a density
24  estimate is produced
25  of the joint distribution of the pair \code{(x(t-k),x(t))}.
26}
27\item{maxlag}{
28  if \code{lags} is not given, it is assigned the value \code{1:maxlag}
29  (default=1).
30}
31\item{ask}{
32  if \code{ask=TRUE}, the program pauses after each plot, until <Enter>
33  is pressed.
34}
35}
36\value{
37a list of two elements, containing the outcome of the estimation of
38the marginal density and the last bivariate density, as produced by
39\code{\link{sm.density}}.
40}
41\section{Side Effects}{
42plots are produced on the current graphical device.
43}
44\details{
45see Section 7.2 of the reference below.
46}
47\references{
48Bowman, A.W. and Azzalini, A. (1997). \emph{Applied Smoothing Techniques for
49Data Analysis: the Kernel Approach with S-Plus Illustrations.}
50Oxford University Press, Oxford.
51}
52\seealso{
53\code{\link{sm.density}}, \code{\link{sm.autoregression}}
54}
55\examples{
56with(geyser, {
57   sm.ts.pdf(geyser$duration, lags=1:2)
58})
59}
60\keyword{nonparametric}
61\keyword{smooth}
62\keyword{ts}
63% Converted by Sd2Rd version 1.15.
64