arima.diag.
The diagnostics include the autocorrelation function of the residuals,
the standardized residuals, and the portmanteau goodness of fit test statistic.
arima.diag.plot(z, layout=<<see below>>, type="h", ...)
arima.diag,
including the components
std.resid (or
resid),
acf.list, and/or
gof.
z.
"l" and
"p".
std.resid (or
resid).
std.resid is present, a plot of the standardized residuals
will be produced.
Otherwise, if
resid is present, a plot of the raw residuals will be given.
If
acf.list is present, then
acf.plot will be called to produce
a plot of the autocorrelation function of the residuals
with a 95% confidence band.
If
gof is present, then the p-values for the Portmanteau will be plotted.
The plot for
acf.list is always high density,
and the plot for
gof is always a point plot.
The residual plot is controlled by the
type argument.
Box, G. E. P. and Jenkins, G. M. (1976). Time Series Analysis: Forecasting and Control. Holden-Day, Oakland, Calif. Chapter 8.
# compute and plot diagnostics for simulated AR(1) series x <- arima.sim(model=list(ar=.9)) fit <- arima.mle(x,model=list(ar=.9)) diag.fit <- arima.diag(fit) # this is equivalent to diag.fit <- arima.diag(fit,plot=F) arima.diag.plot(diag.fit)