www.cmm.bristol.ac.uk/forum

Thank you so much for all your helpful explanations!

Re your last point, I've got these covariates: age (continuous), education (categorical), household income (categorical), urbanicity (categorical), ...

Of these, I realized that household income only varied at the level 2 (household-level) or ...

This is really helpful and thank you again for the details!

(1) Re the random part estimates for the region (highest level), I checked the data and it appears that the region (n=16) is sorted correctly. It's min=1, max=16, and categorical=true. In the view/edit data window, I can see that obs are ...

Hi Chris,

Thank you for your advice!

I've checked the covariance matrix (in the null model where it works) and it seems the variance estimates were:

by IGLS estimation:
- the highest level (region): 0.000(0.000)
- the second highest (neighborhood): 0.034 (0.016)
- the third highest (family): 1 ...

Thank you for your kind response, Chris!

Then, I just tried to run regular Poisson model (four-level with MCMC, offset=1 for all observations). It seems that the null model works, but when I added covariates, I got error messages as this:

- "covariance matrix is not positive-definite"
- "MCMC ...

Also, I've tried to run just poisson and negative binomial model with the above setting, just by replacing 1 (for normal distribution) with 3 (for poisson) and 8 (for negative binomial), but in both cases, the model automatically switched to normal linear model. Am wondering what additional ...

This looks very helpful and thanks for the useful discussion everyone!

Dear Chris,

Thank you so much again for your wonderful support! It's been really helpful and I appreciate it a lot!

While making my projects move forward, I've got a few quick questions. Briefly, my project is as follows:

(1) outcome: depressive symptoms score (0-27), which almost looks like zero ...

Thank you for your help, Chris! So helpful!

Hello,

Hope this finds you all well! I've found this forum really helpful and useful and thank you for all great supports!

Just got a quick question re a prediction in MCMC model (and in general non-MCMC model too).

For instance, I've built a four-level linear model with MCMC (non-specified prior ...

Thank you very much for all your helps Chris! Will try more and get it back to you again!
Best wishes,
Yongjoo