Hi Stata experts,
I have two problems and need advice.
1. I use gllamm command to run multi-level analyses.
When I weighted the data (level-1 weighting), the
model could not converge. There are around 78000 cases
at level 1 and 23 at level 2. The following are my
commands. Could you help me with specifiying the model
so that it could converge.
Random-intercept model:
gen wt1=wt;
gen cons=1;
eq inter: cons espsodem espcons esplib espcomm
espuncl;
gllamm poor d1..., i (country) nrf (1) eqs (inter)
weight (wt) link (logit) family (binom);
2. My second question:
The original income variable is ordinal variable
(categorized as <10000, 10000-25000,
25000-50000.....). I would like to creat a new income
variable with values as the cumulative percentiles. I
know how to get the percentiles for each category
(egen rank). But is there any command with which I
could assign the percentile to each case (that's,
create a new variable and variable values are
cumulative percentiles)?
Thank you very much for your help.
Chloe
____________________________________________________________________________________
Fussy? Opinionated? Impossible to please? Perfect. Join Yahoo!'s user panel and lay it on us. http://surveylink.yahoo.com/gmrs/yahoo_panel_invite.asp?a=7
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/