问题
I'm running an lme -model from the package nlme on R v.3.0.2.
I'm trying to extract the model estimates with predict.lme, but it returns an error. Here's the code to replicate the error:
my.model = lme(fixed = Maxi ~ Time*Origin, random = ~ 1 |Genotype, method = "REML", weights=varPower(), data=dd)
new.my.model <- data.frame(Origin = c("Ka", "Ka", "La", "La"), Time = c("mor", "eve", "mor", "eve"))
predict(my.model, new.my.model, level = 0:1)
#Error in predict.lme(my.model, new.my.model, level = 0:1) :
#cannot evaluate groups for desired levels on 'newdata'
and here's the data:
dd<-read.table(text="Origin Genotype Time Maxi
Ka Ka1 mor 14,59
Ka Ka1 eve 13,42
Ka Ka11 mor 14,08
Ka Ka11 eve 16,29
Ka Ka15 mor 14,38
Ka Ka15 eve 14,56
La La1 mor 17,82
La La1 eve 13,28
Ka Ka1 mor 16,44
Ka Ka1 eve 15,52
Ka Ka15 mor 13,76
Ka Ka15 eve 13,55
Ka Ka1 mor 19,15
Ka Ka1 eve 19,12
La La6 mor 10,54
La La6 mor 11,38
La La6 eve 10,48
Ka Ka15 mor 15,25
Ka Ka15 eve 16,51
La La1 mor 17,46
La La1 eve 15,57
Ka Ka1 mor 16,83
Ka Ka1 eve 15,63
Ka Ka15 mor 14,54
Ka Ka15 eve 15,09
La La1 mor 11,3
La La1 eve 11,94", header=TRUE, dec=",")
Data is two factors (Ka, La) with two levels each (mor, eve) and a random effect. I'm trying to extract the estimates for the 4 levels of data, i.e. Ka,mor ; Ka,eve ; La,mor ; La,eve.
First asked over at CV in another context, but I was recommended here.
回答1:
If you want to ignore the random component, then setting levels=0:1 is not correct; you should just use levels=0.
new.my.model <- expand.grid(Origin=c("Ka","La"), Time=c("mor","eve"))
( pm<-predict(my.model, new.my.model, level = 0) )
# [1] 15.30408 15.51259 12.99398 12.05721
# attr(,"label")
# [1] "Predicted values"
cbind(new.my.model, pred.mean=pm)
# Origin Time pred.mean
# 1 Ka mor 15.30408
# 2 La mor 12.99398
# 3 Ka eve 15.51259
# 4 La eve 12.05721
来源：https://stackoverflow.com/questions/28684645/r-lme-cannot-evaluate-groups-for-desired-levels-on-newdata