sm.ts.pdf: Nonparametric density estimation of stationary time series...

In sm: Smoothing Methods for Nonparametric Regression and Density Estimation

Nonparametric density estimation of stationary time series data

Description

This function estimates the density function of a time series x, assumed to be stationary. The univariate marginal density is estimated in all cases; bivariate densities of pairs of lagged values are estimated depending on the parameter lags.

Usage

sm.ts.pdf(x, h = hnorm(x), lags, maxlag = 1, ask = TRUE)

Arguments

x	a vector containing a time series
h	bandwidth
lags	for each value, k say, in the vector lags a density estimate is produced of the joint distribution of the pair (x(t-k),x(t)).
maxlag	if lags is not given, it is assigned the value 1:maxlag (default=1).
ask	if ask=TRUE, the program pauses after each plot, until <Enter> is pressed.

Details

see Section 7.2 of the reference below.

Value

a list of two elements, containing the outcome of the estimation of the marginal density and the last bivariate density, as produced by sm.density.

Side Effects

plots are produced on the current graphical device.

References

Bowman, A.W. and Azzalini, A. (1997). Applied Smoothing Techniques for Data Analysis: the Kernel Approach with S-Plus Illustrations. Oxford University Press, Oxford.

See Also

sm.density, sm.autoregression

Examples

with(geyser, {
   sm.ts.pdf(geyser$duration, lags=1:2)
})

sm documentation built on July 4, 2022, 5:06 p.m.