1\name{sm.autoregression}
2\alias{sm.autoregression}
3\title{
4Nonparametric estimation of the autoregression function
5}
6\description{
7This function estimates nonparametrically the autoregression function
8(conditional mean given the past values) of a time series \code{x},
9assumed to be stationary.
10}
11\usage{
12sm.autoregression(x, h = hnorm(x), d = 1, maxlag = d, lags,
13                  se = FALSE, ask = TRUE)
14}
15\arguments{
16\item{x}{
17vector containing the time series values.
18}
19\item{h}{
20the bandwidth used for kernel smoothing.
21}
22\item{d}{
23number of past observations used for conditioning; it must be 1
24(default value) or 2.
25}
26\item{maxlag}{
27maximum of the lagged values to be considered (default value is \code{d}).
28}
29\item{lags}{
30if \code{d==1}, this is a vector containing the lags considered for conditioning;
31if \code{d==2}, this is a matrix with two columns, whose rows contains pair of
32values considered for conditioning.
33}
34\item{se}{
35if \code{se==T}, pointwise confidence bands are computed of approximate level 95\%.
36}
37\item{ask}{
38if \code{ask==TRUE}, the program pauses after each plot until <Enter> is pressed.
39}
40}
41\value{
42a list with the outcome of the final estimation (corresponding to
43the last value or pairs of values of lags), as returned by \code{sm.regression}.
44}
45\section{Side Effects}{
46graphical output is produced on the current device.
47}
48\details{
49see Section 7.3 of the reference below.
50}
51\references{
52Bowman, A.W. and Azzalini, A. (1997).
53\emph{Applied Smoothing Techniques for Data Analysis: }
54\emph{the Kernel Approach with S-Plus Illustrations.}
55Oxford University Press, Oxford.
56}
57\seealso{
58\code{\link{sm.regression}}, \code{\link{sm.ts.pdf}}
59}
60\examples{
61sm.autoregression(log(lynx), maxlag=3, se=TRUE)
62sm.autoregression(log(lynx), lags=cbind(2:3,4:5))
63}
64\keyword{nonparametric}
65\keyword{smooth}
66\keyword{ts}
67% Converted by Sd2Rd version 1.15.
68