On 17/02/2009, at 10:54 AM, dimitris kapetanakis wrote:
Dear all,
I am trying to estimate the prediction from a fixed effects model
and their
confidence intervals as well. Though I do not want to include in the
prediction and at the confidence intervals the intercept. For that
reason I
used the argument incl.non.slopes=FALSE. But either if it is TRUE
or FALSE
it does not have any difference and also the system does not
provide any
warning. I really cannot understand what is happening and I use
both predict
and predict.lm but there is no difference.
Explicitly the code is:
fe.nox <- lm(nox~ state.1 + state.2 + state.3 + state.4 + state.
5 +
state.6 + state.7 + state.8 + state.9 + time.1 + time.2 +
time.3 +
time.4 + time.5 + time.6 + time.7 + pcinc + I(pcinc^2) + I
(pcinc^3),
data=ekc)
p.fe.nox<-predict.lm(fe.nox, new, interval = "prediction", level=0.95,
incl.non.slopes=FALSE)
Any Help would be highly appreciated
Where do you get a predict.lm() function that has an argument
``incl.non.slopes''??? Neither the help file nor
args(predict.lm) reveal any trace of such an argument.
You must be using a modified version of this function. So check
with whomever you got it from as to why this argument is not having
the effect you expect. It is not at all clear to me what you *do*
expect. Are you trying, artificially, to set the intercept to 0
before predicting? Why would you want to do that?
Of course you don't get any difference between predict() and
predict.lm().
The predict() function is generic, and fe.nox is of class "lm" so the
method predict.lm() will be used.
I'm sure your data could be structured so that your model could be
written with much less verbosity.
cheers,
Rolf Turner
######################################################################
Attention:\ This e-mail message is privileged and confid...{{dropped:9}}
______________________________________________
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
