[R-sig-ME] Formula and Start-values … nlmer (actual error messages)
Robert Chatfield
chatfield at alumni.rice.edu
Mon Jan 13 18:32:44 CET 2014
Hmm: I should have included the actual error message I received from nlmer
at about the line "Could this naming be confusing to nlmer? The start values must be named."
> first.nlmer = nlmer( y ~ a*(x+ x.nought) ~ (a|species.type) + (x.nought|id),
+ start=start.vals.1, data=good.est.merge.long.dat )
Error in nlmer(y ~ a * (x + x.nought) ~ (a | species.type) + (x.nought | :
not all parameter names are used in the nonlinear model expression
where named list is given by ...
> start.vals.1 = c(a=aij, x.nought=C.nought.lin)
> start.vals.1[1]
a1
0.002266363
> start.vals.1[3604]
x.nought1802
-0.1204595
Using single values with a single name also does not work, nor does a construct
with cbind:
> first.nlmer = nlmer( y ~ a*(x+ x.nought) ~ (a|species.type) + (x.nought|id),
+ start=c(a=0.00364901,x.nought=0), data=good.est.merge.long.dat )
Error in nlmer(y ~ a * (x + x.nought) ~ (a | species.type) + (x.nought | :
gradient attribute of evaluated model must be a numeric matrix
Bob Chatfield
More information about the R-sig-mixed-models
mailing list