cable.ar.p.diag: Bent-Cable AR(p>0) Diagnostics

Description

ACF, PACF, and other plots are produced for diagnosing an AR(p) bent-cable fit when p>0.

Usage

cable.ar.p.diag(ar.p.fit, resid.type = "p", xlab = "time", ylab = "",
	main = NULL, main.all = NULL, ctp.ci = NULL)

Arguments

ar.p.fit

A cable.ar.p.iter object for AR(p) data, p>0.

resid.type

A type argument for the plot function; used to control the way fitted residuals and innovations are displayed. Default is "p" for points.

xlab

Character string: x-axis label.

ylab

Character string: y-axis label.

main

Character string: title for the time series plot.

main.all

Character string: title for the entire set of plots.

ctp.ci

A cable.change.conf object.

Warning

See the warnings from cable.ar.p.plot and cable.ar.p.resid.

Details

This function splits the plotting canvas into several panels. For one panel, ar.p.fit is fed to cable.ar.p.plot that produces a scatterplot of the data and overlays on it the fitted bent cable with the estimated transition. The optioinal ctp.ci is also fed to cable.ar.p.plot to add the CTP confidence interval to the same panel. Additionally, ar.p.fit is fed to cable.ar.p.resid to extract the fitted residuals and innovations, which are then plotted in separate panels that again show the estimated transition and confidence interval. Finally, four panels show ACF and PACF diagnostics for the fitted residuals and innovations, via the built-in R functions acf and pacf.

References

See the bentcableAR package references.

Examples

Run this code

data(sockeye)

# AR(2) cable fit
fit.ar2 <- cable.ar.p.iter( c(13,.1,-.5,11,4,.5,-.5),
	sockeye$logReturns, tol=1e-4 )
cable.ar.p.diag( fit.ar2, main="bent cable", main.all="Sockeye",
	ctp.ci=cable.change.conf( fit.ar2, .9 ) )
		# compare to this:
		# fit.ar2 <- bentcable.ar( sockeye$logReturns,
		#	init.cable=c(13,.1,-.5,11,4), p=2, main="Sockeye bent cable",
		#	ci.level=.9 )

# AR(4) stick fit
fit.ar4 <- cable.ar.p.iter( c(13,.1,-.5,11,.5,-.5,.5,-.5),
	sockeye$logReturns, tol=1e-4, stick=TRUE )
cable.ar.p.diag( fit.ar4, ctp.ci=cable.change.conf( fit.ar4, .95 ) )
		# compare to this:
		# fit.ar4 <- bentcable.ar( sockeye$logReturns,
		#	init.cable=c(13,.1,-.5,11), p=4, stick=TRUE )

Run the code above in your browser using DataLab

Get 50% off unlimited learning