ellipse(center, shape, radius, log="", center.pch=19, center.cex=1.5,
segments=51, add=TRUE, xlab="", ylab="",
col=palette()[2], lwd=2, grid=TRUE, ...)
dataEllipse(x, y, log="", levels=c(0.5, 0.95), center.pch=19, center.cex=1.5,
plot.points=TRUE, add=!plot.points, segments=51, robust=FALSE,
xlab=deparse(substitute(x)),
ylab=deparse(substitute(y)),
col=palette()[1:2], lwd=2, grid=TRUE, ...)
confidenceEllipse(model, ...)
## S3 method for class 'lm':
confidenceEllipse(model, which.coef, levels=0.95, Scheffe=FALSE,
center.pch=19, center.cex=1.5, segments=51, xlab, ylab,
col=palette()[2], lwd=2, add=FALSE, ...)
## S3 method for class 'glm':
confidenceEllipse(model, which.coef, levels=0.95, Scheffe=FALSE,
center.pch=19, center.cex=1.5, segments=51, xlab, ylab,
col=palette()[2], lwd=2, add=FALSE, ...)
"x"
if the x-axis is logged, "y"
if the y-axis is
logged, and "xy"
TRUE
add ellipse to current plot.y
is missing) a 2-column numeric matrix.x
.FALSE
data ellipses are added to the current scatterplot,
but points are not plotted.TRUE
use the cov.trob
function in the lm
or glm
.TRUE
scale the ellipse so that its projections onto the
axes give Scheffe confidence intervals for the coefficients.2
(see par
).plot
and
line
.NULL
. These functions are used for their side effect: producing
plots.dataEllipse
superimposes the normal-probability contours over a scatterplot
of the data.cov.trob
.dataEllipse(Prestige$income, Prestige$education, levels=0.1*1:9, lty=2)
confidenceEllipse(lm(prestige~income+education, data=Prestige), Scheffe=TRUE)
Run the code above in your browser using DataLab