I have a linear mixed model lme (from package nlme) with distance (log 10 transformed) as the response variable, a factor variable pre or post translocation as a predictor, time log10 transformed as an offset fixed effect, and then two nested random effects ID (names) nested within population (factor with two levels) and then sex (male or female) nested within season (breeding or non-breeding)
m.lme2 <- lme(l.S_Dists ~ PrePost_Trans + offset(l.St), random=list(Population=pdDiag(~ ID), Season= pdDiag(~Sex)), data = Q1a)
I run this fine - and check AIC which is 3562.669
Everything so far is fine.
I then go to check this AIC with the null model to see if m.lme2 fares better. Using the code:
m.lme0 <- lme(l.S_Dists ~ 1 + offset(l.St), random=list(Population=pdDiag(~ ID), Season= pdDiag(~Sex)), data = Q1a)
However, when I try and run this model I get the error message:
Error in lme.formula(l.S_Dists ~ 1 + offset(l.St), random = list(Population = pdDiag(~ID), : nlminb problem, convergence error code = 1 message = singular convergence (7)
I have run this model previously for a very similar data set and get no error.
Can somebody please tell me why I would be getting this error message when I try and run the null model?
Here's what the summary looks like for the data set Q1a. Have uploaded image links of this. Summary of the data set Q1a. Summary of the data set Q1a
