[R] nlme predict with se?

Martin Maechler maechler at stat.math.ethz.ch
Tue Mar 21 09:51:59 CET 2006

>>>>> "Emilio" == Emilio A Laca <ealaca at ucdavis.edu>
>>>>>     on Sat, 18 Mar 2006 22:01:05 -0800 writes:

    Emilio> Berton, What you say makes sense and helps. I could
    Emilio> do a parametric bootstrapping. However, I cannot
    Emilio> make predict.nlme work at all, even without the
    Emilio> se's. It would save me a lot of time to be able to
    Emilio> use the predict as the statistic in boot.

    Emilio> Does "predict.nlme" work at all? 
    Emilio> It is a listed method in MEMSS. Is there an example?  

  help(predict.nlme)       or even more directly

show that it does work (at least in some cases).
If you read the help page, you can also deduce that an
argument 'se' is not made use of.

Please do give us a *reproducible* (for us!) example as you are
asked to do by the posting guide. 
For this you must use an R or nlme example data set; or
artificially generated data for which you show the exact
generative statements {including a set.seed(*) one if you use
random number}.

Martin Maechler, ETH Zurich

More information about the R-help mailing list