On Wed, 2006-08-09 at 15:04 -0500, Douglas Bates wrote: > On 8/9/06, Rick Bilonick <[EMAIL PROTECTED]> wrote: > > I'm fitting a mixed effects model: > > > > fit.1 <- lme(y~x,random=~1|id,data=df) > > > > There are two different observations for each id for both x and y. When > > I use plot(fit.1), there is a strong increasing linear trend in the > > residuals versus the fitted values (with no outliers). This also happens > > if I use random=~x|id. Am I specifying something incorrectly? > > Could you provide a reproducible example please? > > I suspect that the problem comes from having only two observations per > level of id. When you have very few observations per group the roles > of the random effect and the per-observation noise term in explaining > the variation become confounded. However, I can't check if this is > the case without looking at some data and model fits.
Unfortunately, I can't send the actual data. I did make a simple intercepts-only example with two observations per group but it does not exhibit the linear trend. library(nlme) x <- rnorm(20,5,1) id <- factor(rep(1:20,each=2)) y <- as.vector(sapply(x,rnorm,n=2,sd=0.2)) df <- data.frame(id,y) df.gd <- groupedData(y~x|id,data=df) summary(lme.1 <- lme(y~1,random=~1|id,data=df.gd)) plot(lme.1) If I fit an intercepts-only model to the actual data, I still see the trend in the residuals. What other analysis would you suggest? Rick B. ______________________________________________ [email protected] 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.
