In article <ifado.list.r.help/Pine.LNX.4.31.0208191041401.13359-100000 at gannet.stats>,
<ripley at stats.ox.ac.uk> wrote:
On Mon, 19 Aug 2002 iwhite at staffmail.ed.ac.uk wrote:
In the simple model Y = mu + U + e, where U and e have variances Vu and
Ve, the BLUP for a subject with observation Y is
E(U|Y) = Vu * (Y - mu)
----
(Vu+Ve)
Similarly for more complicated models, i.e. the BLUP is a simple function
of the variance components. I'm not sure whether this answers the original
question, but it must be relevant and nobody has mentioned it so far.
Yes, it is relevant. In general the BLUP is not a simple function like that, not with hierarchical designs and e.g. random effects on slopes as well as intercepts. Which is why I'd like my software to do it for me.
Let me sharpen my original question. Given new observations (on a new or even an old subject), is there a simple test whether these new observations are fitted by the model? As I understand it, if I had the BLUPs I could just compare the residuals of the fit with the model's residuals and those of a fixed effects model only for the new data. wbk -.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.- r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html Send "info", "help", or "[un]subscribe" (in the "body", not the subject !) To: r-help-request at stat.math.ethz.ch _._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._