FitAR (version 1.94)

RacfPlot: Residual Autocorrelation Plot

Description

Residual autocorrelation plot for "FitAR" objects. This plot is useful for diagnostic checking models fit with the function FitAR.

Usage

RacfPlot(obj, lag.max = 1000, SquaredQ=FALSE, ylab="")

Arguments

obj

output from FitAR

lag.max

maximum lag. Set to 1000 since minimum of this value and the value in the obj is used.

SquaredQ

default is FALSE. For squared residual autocorrelations, set to TRUE

ylab

y-axis label

Value

Plot is produced as a side-effect. No output

Details

The standard deviations of the residual autocorrelations are obtained from McLeod (1978, eqn.16) or McLeod and Zhang (2006, eqn.16). Simultaneous confidence bounds are shown and constructed using the Bonferonni approximation as suggested by Hosking and Ravishanker (1993)

References

Hosking, J.R.M. and Ravishanker, N. (1993) Approximate simultaneous significance intervals for residual autocorrelations of autoregressive-moving average time series models. Journal of Time Series Analysis 14, 19-26.

McLeod, A.I. (1978), On the distribution and applications of residual autocorrelations in Box-Jenkins modelling, Journal of the Royal Statistical Society B 40, 296-302.

McLeod, A.I. and Zhang, Y. (2006). Partial autocorrelation parameterization for subset autoregression. Journal of Time Series Analysis, 27, 599-612.

See Also

plot.FitAR, FitAR,

Examples

#fit subset AR and plot diagnostic check
 data(SeriesA)
 out<-FitAR(SeriesA, c(1,2,7), ARModel="ARp")
 RacfPlot(out)
#note that plot produces LBQPlot and RacfPlot
 plot(out)
#check squared residuals
 RacfPlot(out, SquaredQ=TRUE)