xtmerlogit

[Oliva Safari]

Friends,


I had this query; I am fitting three levels logistic-mixed effect model with Random coefficient variable for both groups. But eventually i cannot see levels of random coefficient variable in the STATA output. Does any one have idea? help me please

this was the command

xtmelogit mc i.age_cat i.parity i.educ_wome i.v190 i.edu_hus i.media i.emp || v101:v190 || v001:v190, variance collinear or

My random coefficient variable is v190. This is dummy-categorical variable with 5-levels, that i was looking for their values.


Thanks a lots
Oliva

[Phil Bromiley]

Welcome to Statalist. You'll increase your chances of a useful answer by following the FAQ on asking questions - provide Stata code in code delimiters, readable Stata output, and sample data using dataex. Knowing what Stata is really producing would really help us help you.

You don't say what Stata version you use. xtmelogit was withdrawn as of Stata 13. It is now available as meqrlogit. I don't use mixed models, but just glancing at your code, I wonder if there is a problem with having i.v190 appear as a variable, and in both random effects statements. I note you include collinear which may suggest you do have a problem there.

[Oliva Safari]

Hi Phil,
Thanks so much for reply, highly appreciated.

I am using STATA V13. The v190 is categorical- factor variable. My understand is i.var is used for factor variables.Would you recommend not to use "collinear" as it allow for collinearity?

Thanks,
Oliva

[Oliva Safari]

I am happy that i have get a way to fit random slope/coefficient with STATA. need to create indicator variable