[R-sig-ME] factor specific residual variance for random coefficient model with lmer

[li li]

Ok. I found from R help page that the weights argument could
accomplish different residual varainces for different factor levels
like below.


fm2 <- lmer( dv ~ time.num*drug + (0+ drug + time.num | Patient ),
data=dat.new,
weights = varIdent(form = ~1 | drug))
summary(fm2)


But the following error term returned.


Error in summary(fm2) :
  error in evaluating the argument 'object' in selecting a method for
function 'summary': Error: object 'fm2' not found

Any advice?

Thanks
  Hanna

2015-06-10 21:14 GMT-04:00, li li <hannah.hlx at gmail.com>:
> Hi all,
>   I am wondering how to specify the model fm1 below so that the two
> groups (treatment and control) specified by the column drug in the
> data matrix have different residual variances. Any suggestion?
>   Please see the codes below.
>   Thanks very much!
>     Hanna
>
>
>
>
> set.seed(500)
> n.timepoints <- 8
> n.subj.per.tx <- 20
> sd.d <- 5;
> sd.p <- 2;
> sd.res <- 1.3
> drug <- factor(rep(c("D", "P"), each = n.timepoints, times =
> n.subj.per.tx))
> drug.baseline <- rep( c(0,5), each=n.timepoints, times=n.subj.per.tx )
> Patient <- rep(1:(n.subj.per.tx*2), each = n.timepoints)
> Patient.baseline <- rep( rnorm( n.subj.per.tx*2, sd=c(sd.d, sd.p) ),
> each=n.timepoints )
> time <- factor(paste("Time-", rep(1:n.timepoints, n.subj.per.tx*2),
> sep=""))
> time.baseline <-
> rep(1:n.timepoints,n.subj.per.tx*2)*as.numeric(drug=="D")
> dv <- rnorm( n.subj.per.tx*n.timepoints*2,
> mean=time.baseline+Patient.baseline+drug.baseline, sd=sd.res )
> dat.new <- data.frame(time, drug, dv, Patient)
> dat.new$time.num = rep(1:n.timepoints, n.subj.per.tx*2)
>
> library(lme4)
> fm1 <- lmer( dv ~ time.num*drug + (0+ drug + time.num | Patient ),
> data=dat.new )
> summary(fm1)
> resid(fm1)
> plot(resid(fm1))
>