Sara Krause <skkrause <at> gmail.com> writes:
>
> Hi,
>
> I am new to R and not fantastic at statistics so it may well be that I am
> doing something silly but I can't figure out what it is and hoping that
> somebody can help.
>
> I am running package lme4, and trying to get a Residuals vs. Fitted graph.
> When I try to plot, I receive an error.
>
> Error in as.double(y) :
> cannot coerce type 'S4' to vector of type 'double'
>
> Here is the code I am using
>
> library("lme4")
>
> m1<-lmer(y~trt*time*loc_block*Season+(1|ID), data=d)
>
> plot(m1,add.smooth=F,which=1)
>
> Error in as.double(y) :
> cannot coerce type 'S4' to vector of type 'double'
>
> I searched the forums for some answers but haven't found anything
> useful. Any insight into what I am doing wrong and how to fix it.
>
> Sara
The standard diagnostics which you have seen in 'lm' are not
implemented for 'lme4'. You have to do it yourself, but it's
not *too* hard:
library(lme4)
## from ?lmer
fm1 <- lmer(Reaction ~ Days + (Days|Subject), sleepstudy)
f <- fitted(fm1)
r <- residuals(fm1)
plot(f,r)
sm <- loess(r~f)
v <- seq(min(f),max(f),length=101)
lines(v,predict(sm,data.frame(f=v)),col=2)
Further questions about lme4 should probably go to the
r-sig-mixed-models mailing list.