[R-sig-ME] lme4/nlmer problem

Kalle Eerikäinen kalle.eerikainen at metla.fi
Fri Aug 17 10:52:23 CEST 2007


I have done some test estimations using the lme4 package of R. The 
estimation of the parameters of linear mixed-effect models using the 
'lmer' goes well. For instance, a very simple single-level mixed model 
"Naslund_lmer1  <- lmer(y ~ d13 + (1 |stand), data = height1)" for the 
relationship between the tree height (or its transformation) and 
diameter comes out perfectly.

However, if I attempt to estimate nonlinear mixed-effect models, I 
always receive an error message that tells me: "Error: 
length(start$fixed) is not TRUE". This is also the case with the 
following model:

Schumacher_nlmer1 <- lme4:::nlmer(ht   ~ exp(p0 + p1*1/d13)+u0 ~ 
(u0|stand) , fixed=p0+p1~1, data = height1, start = c(p0 = 0.1, p1 = 
-9.0), verb = 1)

It is obvious that there exists a trivial bug/mistake in my code, but I 
cannot see what is wrong with it. Could someone give me a helping hand 
with my 'nlmer problem'?


Kalle Eerikäinen

More information about the R-sig-mixed-models mailing list