[R-sig-ME] Warning message: In mer_finalize(ans) : gr cannot be computed at initial par (65)

[David Duffy]

On Wed, 16 Feb 2011, Kris Jones wrote:

> When I run the model with just one fixed effect in the model (Volume 
> sampled), the model works fine. However, when I plug in other variables, 
> I get the error 'Warning message: In mer_finalize(ans) : gr cannot be 
> computed at initial par (65)'. I'm not quite sure what this means, and 
> what I should do to deal with this error. There are a lot of NAs in the 
> data, so I'm not sure if that is contributing to this issue. I haven't 
> had much luck troubleshooting this problem, so any help you can provide 
> would be much appreciated.

Your dataset doesn't seem to have come through, but my guess is that
there are not enough data (given your NAs) to allow fitting of such an 
elaborate model.  Does a "plain" ZIP regression

Juv_Catch ~ Volume + Secchi + EcTop + WaterTemp + AvgD_2 + HoleD_2 + 
Channel_Wdth + Velocity2 + Year

run, and give sensible answers?

Do

glmer(Juv_Catch ~ Volume + Secchi + Year + (1|Station2) + (1|obs)
glmer(Juv_Catch ~ Volume + Secchi + (Year|Station2) + (1|obs)
etc

run?

Cheers, David Duffy.

-- 
| David Duffy (MBBS PhD)                                         ,-_|\
| email: davidD at qimr.edu.au  ph: INT+61+7+3362-0217 fax: -0101  /     *
| Epidemiology Unit, Queensland Institute of Medical Research   \_,-._/
| 300 Herston Rd, Brisbane, Queensland 4029, Australia  GPG 4D0B994A v