Statalist The Stata Listserver


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: gllamm converge problem/cumulative percentiles


From   Tsui-o Tai <[email protected]>
To   [email protected]
Subject   st: gllamm converge problem/cumulative percentiles
Date   Mon, 25 Jun 2007 22:22:21 -0700 (PDT)

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/



© Copyright 1996–2024 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index