Displaying 1 result from an estimated 1 matches for "nohet".
Did you mean:
nodet
2010 Mar 22
0
using lmer weights argument to represent heteroskedasticity
...dev),
as.numeric(attr(VarCorr(o), "sc")))
names(r) <- c("b0","b0sd","b1","b1sd","sdgroup","sdresid")
return(r)
}
res <- vector(0, mode="list")
library(nlme)
## lme, ignoring heteroskedasticity
res[["lme.nohet"]] <- sumLME( lme(fixed = y ~ x, data = d, random = ~1
| groupid) )
## lme, heteroskedastic model with variance weights estimated
res[["lme.esthet"]] <- sumLME( lme(fixed = y ~ x, data = d, random =
~1 | groupid, weights = varIdent(form = ~1|verrorf)) )
## lme, heteroskedasti...