Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

st: Problems with latent class analysis using gllamm


From   Ijon Tichy <ijontichy7@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   st: Problems with latent class analysis using gllamm
Date   Sun, 22 Mar 2009 23:24:42 +0100

Dear Statalisters,

I tried to run a latent class analysis using Stata 10. I have the
following manifest, numeric variables:

id – respondent’s identification
country – identifying respondent’s country of origin
occupationfatherworking – a categorical variable containing
occupational categories that describe the job of respondent’s father
fisced3completed, fisced4completed, fisced5bcompleted,
fisced5acompleted – four dummy variables measuring the level of
education achieved by respondent’s father
wealthindex – an index measuring wealth of household

I expanded my dataset in the following way:

eq classfather: fiscoworking1-fiscoworking533 fisced3completed
fisced4completed fisced5bcompleted fisced5acompleted
wealthindex4dum1-wealthindex4dum4

{fiscoworking1-fiscoworking533 and wealthindex4dum1-wealthindex4dum4
are dummy variables created from occupationfatherworking and
wealthindex}

Then I specified:

eq latentfather: occupationfatherworking  fisced3completed
fisced4completed fisced5bcompleted fisced5acompleted wealthindex4cats

and – to get the full independence model – I run:

gllamm classfather, i(id country1) eq(classfather) geq(latentfather)
nrf(533) ip(f) nip(1) adapt

To which Stata answered:

variable classfather not found
r(111);

Could anyone point me out where did I make a mistake, please?

Best regards,
Ijon

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index