| linesHyperb.lm {sfsmisc} | R Documentation |
Add confidence/prediction hyperbolas for y(x_0) to a plot with data or regression line.
linesHyperb.lm(object, c.prob=0.95, confidence=FALSE,
k=if (confidence) Inf else 1,
col=2, lty=2, do.abline=TRUE)
object |
result of lm(.). |
c.prob |
coverage probability in (0,1). |
confidence |
logical; if true, do (small) confidence band, else,
realistic prediction band for the mean of k observations. |
k |
integer or Inf; assume k future observations;
k = Inf corresponds to confidence intervals (for y). |
col, lty |
attributes for the lines to be drawn. |
do.abline |
logical; if true, the regression line is drawn as well. |
With predict.lm(*, interval=) is available,
this function linesHyperb.lm is only slightly more general for
its k argument.
Martin Maechler, Oct 1995
predict.lm(*, interval=) optionally computes
prediction or confidence intervals.
data(swiss)
plot(Fertility ~ Education, data = swiss) # the data
(lmS <- lm(Fertility ~ Education, data = swiss))
linesHyperb.lm(lmS)
linesHyperb.lm(lmS, conf=TRUE, col="blue")