animation (version 1.1-4)

mwar.ani: Demonstration for ``Moving Window Auto-Regression''...

Description

Demonstration for ``Moving Window Auto-Regression'' This function just fulfills a very naive idea about moving window regression using rectangles to denote the ``windows'' and move them, and the corresponding AR(1) coefficients as long as rough confidence intervals are computed for data points inside the ``windows'' during the process of moving.

Usage

mwar.ani(x, k=15, conf=2, mat=matrix(1:2, 2), widths=rep(1, ncol(mat)),
    heights=rep(1, nrow(mat)), lty.rect=2, ...)

Arguments

x

univariate time-series (a single numerical vector); default to be sin(seq(0, 2 * pi, length = 50)) + rnorm(50, sd = 0.2)

k

an integer of the window width

conf

a positive number: the confidence intervals are computed as [ar1 - conf*s.e., ar1 + conf*s.e.]

mat,widths,heights

arguments passed to layout to divide the device into 2 parts

lty.rect

the line type of the rectangles respresenting the moving ``windows''

...

other arguments passed to points in the bottom plot (the centers of the arrows)

Value

• A list containing
• phi
  the AR(1) coefficients
• L
  lower bound of the confidence interval
• U
  upper bound of the confidence interval

Details

The AR(1) coefficients are computed by arima.

References

Robert A. Meyer, Jr. Estimating coefficients that change over time. International Economic Review, 13(3):705-710, 1972. http://animation.yihui.name/ts:moving_window_ar

See Also

arima

Examples

# moving window along a sin curve 
oopt = ani.options(interval = 0.1, nmax = 50)
op = par(mar = c(2, 3, 1, 0.5), mgp = c(1.5, 0.5, 0))
mwar.ani(lty.rect = 3, pch = 21, col = "red", bg = "yellow",type='o')

# for the data 'pageview'
ani.options(interval = 0.1, nmax = 50)
data(pageview)
mwar.ani(pageview$visits, k = 30)

par(op)

# HTML animation page
ani.options(ani.height = 500, ani.width = 600, nmax = 50,
title = "Demonstration of Moving Window Auto-Regression",
description = "Compute the AR(1) coefficient for the data in the
window and plot the confidence intervals. Repeat this step as the
window moves.")
ani.start()
par(mar = c(2, 3, 1, 0.5), mgp = c(1.5, 0.5, 0))
mwar.ani(lty.rect = 3, pch = 21, col = "red", bg = "yellow",type='o')
ani.stop()

ani.options(oopt)