Creates a set plots useful for comparing fitted linear models stored in a
class fit.models object belonging to virtual class lmfm. The plot options
are (2) Normal QQ-Plot of Residuals, (3) Estimated Kernel Density of
Residuals, (4) Robust Residuals vs Robust Distances, (5) Residuals vs
Fitted Values, (6) Sqrt of abs(Residuals) vs Fitted Values, (7) Response vs
Fitted Values, (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 lmfm.
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).
chisq.percent
p-value used to calculate the chi-squared quantile used as the outlier
threshold for robust distances.
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 in computing 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.
cutoff
if TRUE, bounds are added to the one variable regression plot.
Observations outside these bounds received zero weight in the analysis.
VALUE:
x is invislbly returned.
SIDE EFFECTS:
The selected plots are drawn on a graphics device.
DETAILS:
If all the linear models in the fit.models object have exactly one
explanatory variable then there is the additional option to plot the fit
over a scatter plot of the data.
REFERENCES:
Atkinson, A. C. (1985). Plots, Transformations and Regression. New York:
Oxford University Press.
SEE ALSO:
,
.
EXAMPLES:
stack.lm <- lm(Loss ~ ., data=stack.dat)
stack.lmRob <- lmRob(Loss ~ ., data=stack.dat)
stack.fm <- fit.models(stack.lmRob, stack.lm)
#
# stack.fm is a fit.models object with virtual class lmfm.
#
plot(stack.fm)