plot.arrob: Diagnostic Plots for a Fitted Autoregressive Model

In robts: Robust Time Series Analysis

Description

Diagnostic plots for objects of class "arrob" as it is returned by a call to the function arrob.

Usage

## S3 method for class 'arrob'
plot(x, ask = TRUE, ci = 0.95, ...)

Arguments

x	an object of class "arrob". Usually the result of a call to arrob.
ask	logical value. If TRUE (and the R session is interactive) the user is asked for input, before a new figure is drawn (see devAskNewPage).
ci	coverage probability for confidence intervals in some of the plots. Plotting of the confidence interval is suppressed if ci is zero or negative.
...	further arguments are currently ignored. Only for compatibility with generic function.

Details

Produces the following plots:

• Information criterion for different model orders. For robust fits obtained by arrob the plot shows a robust information criterion; for classical fits obtained by ar it shows differences of Akaikes Information criterion (AIC) with the AIC of the best model. The best model is marked by a vertical line. This plot is helpful for selection of the optimal order of an AR model.

• Robust autocorrelation function of the residuals. The autocorrelation function is robustly estimated using the default settings of acfrob. If the confidence level ci is greater than zero, then confidence limits under the assumption of uncorrelated residuals are added to the plot, see plot.acfrob. This plot is helpful to check whether the residuals are uncorrelated.

• Residuals plotted against time. If the confidence level ci is greater than zero, then confidence limits under the assumption of uncorrelated residuals with zero mean and constant variance are added to the plot. This plot is helful to check whether the residuals have mean zero and constant variance and to identify possible outliers.

• Normal quantile-quantile plot of the residuals, see functions qqnorm and qqline. This plot is helpful to check whether the residuals are from a normal distribution.

Author(s)

Alexander Dürre and Tobias Liboschik

See Also

arrob for robust fitting of AR models, ar for classical fitting.

Other S3 methods residuals.arrob, fitted.arrob, filtered.arrob and predict.arrob for objects of class "arrob".

Examples

set.seed(1066)
tss <- arima.sim(model = list(ar = 0.3, ma = 0.5), n = 100)
fit <- arrob(tss, method = "yw")
plot(fit)

robts documentation built on May 2, 2019, 4:55 p.m.