Creates a set plots useful for comparing fitted analysis of variance models
stored in a class fit.models object belonging to virtual class
aovfm. The plot options are (2) Normal QQ-Plot of Residuals,
(3) Estimated Kernel Density of Residuals, (4) Residuals vs Fitted
Values, (5) Sqrt of abs(Residuals) vs Fitted Values, (6) Response vs
Fitted Values, (7) Residual-Fit Spread, (8) Standardized Residuals vs
Index (Time), (9) Overlaid Normal QQ-Plot of Residuals, and (10)
Overlaid Estimated Density of Residuals.
a fit.models object belonging to virtual class aovfm.
OPTIONAL ARGUMENTS:
which.plots
either "ask", "all", or an integer vector specifying which plots to
draw. If which.plots is an integer vector, use the plot numbers
given in the description above (or in the "ask" menu).
vertical.outlier
p-value used to calculate the standard normal quantile used as the outlier
threshold for residuals.
smooths
if TRUE, smooth curves are approximated to the scatterplots using
loess.smooth and added to the appropriate plots.
rugplot
if TRUE, a univariate histogram or rugplot is displayed along the base of
each plot, showing the occurrence of each x-value; ties are broken by
jittering.
id.n
number of outliers identified in plots.
envelope
if TRUE, a simulation envelope is added to the QQ-plot.
half.normal
if TRUE, half normal QQ-plots will be used.
robustQQline
if TRUE, a robust fit is added to the QQ-plot.
mc.samples
number of samples used to compute the simulation envelope.
level
confidence level for the simulation envelope.
seed
an integer between 0 and 1023. The seed value used for random number
generation in the QQ-plot simulation envelope.
VALUE:
x is invisibly returned.
SIDE EFFECTS:
The selected plots are drawn on a graphics device.
REFERENCES:
Atkinson, A. C. (1985). Plots, Transformations and Regression. New York:
Oxford University Press.
SEE ALSO:
,
.
EXAMPLES:
lawson.aov <- aov(Loss ~ ., data=lawson.dat)
lawson.aovRob <- aovRob(Loss ~ ., data=lawson.dat)
# lawson.fm is a fit.models object with virtual class aovfm:
lawson.fm <- fit.models(lawson.aovRob, lawson.aov)
plot(lawson.fm)