Hi, Another student here got the problem figured out! R was calling the variable "r" a factor, not a number, and then it couldn't fit an lme to a factor response variable Thank you! Megan Perry Alaska Cooperative Fish and Wildlife Research Unit - USGS Institute of Arctic Biology 211 Irving I PO Box 757020 University of Alaska Fairbanks Fairbanks, AK 99775-7020 Office: 907-474-1133 Lab: 907-474-6740 m.perry at uaf.edu
Ben Bolker wrote:
this isn't a reproducible example. Can you pull out a small subsample of the NPLW data set that gives the same error, dump it with dput(), and send to the list so that we can see what's going on? Short of that, str(NPLW) *might* give us more of a hint. Ben Bolker Megan Perry wrote:
Hi, I am trying to analyze my data using a linear mixed effects model, using the nlme package. The model I am using is:
>LW<-lme(fixed=r ~ t,random = ~ 1 | s, method=c("ML"), data=NPLW)
but, I am stumped with the error message:
Error in `rownames<-`(`*tmp*`, value = c("1", "2", "3", "4", "5", "6", :
attempt to set rownames on object with no dimensions
In addition: Warning message:
In Ops.factor(y[revOrder], Fitted) : - not meaningful for factors
Any help would be really appreciated!!
Thanks,
Megan