As I understand it, predict.lm(l ,newdata=nd ,interval="confidence")
yields confidence bands for the predicted mean of new observations and
lm.predict(l ,newdata=nd ,interval="prediction") yields confidence
bands for new observations themselves, given an lm object l.
However with regard to {mgcv} although predict.gam (g ,se.fit=TRUE ,interval=
"prediction") computes without protest it only seems to yield the
predicted mean and the standard error of the predicted mean given an {mgcv} gam
object g.
Is there an easy way to get confidence bands for new observations for {mgcv} gam
objects, analogous to those available for lm objects?
[[alternative HTML version deleted]]
