Re: st: question on gllamm with ip(f), the discrete latent variable

[Stas Kolenikov <skolenik@gmail.com>]

That's an odd error message. Make sure your copy of gllamm is
installed correctly; type -ssc install gllamm, replace-.

On Sun, May 2, 2010 at 1:26 PM, Hey Sky <heyskywalker@yahoo.com> wrote:
>
> hey, All
>
> I am new to gllamm and have read the manual and some other documents for it,
> such as A review of random effects modelling using gllamm in Stata by Grilli, L.
> and Rampichini on gllamm.org website.
>
> I am modeling people training behavior base on their formal education history, ie
> ed9, acedu, etc, and accumulative on/off job training history, ie acojt, acoft, etc
>  with panel data. I assume there is one latent variable, epsilon, with two types.
> thus every individual  has a probability belong to each type. and I have code like
> below, which different choice has different explain variables controlled by constr.
>
> now my question is, everytime I try to run the gllamm, it will give me the
> following warning:
>
> "specify at least one file or folder to include in the zip file
> (error occurred in ML computation)
> (use trace option and check correctness of initial model)"
>
> I have tried Grilli, L's code and data set for discrete latent variable, which are given
>  on gllamm.org, but the same warning appears and the program cannot continue.
> the trace does not help or maybe I skiped the message.
>
> I know the question should be a very basic but why is that and how to solve it?
>
> any hints are appreciated and thanks a lot in advance
>
>
>
> code:
> local contr234 "ed9 ed10 ed11 ed12 ed13 ed14 ed15 ed16 ed17"
> constr def 1 [c1]: acedu acexp acojt acoft
> constr def 2 [c2]: `contr234'
> constr def 3 [c3]: `contr234'
> constr def 4 [c4]: `contr234'
>
> gen cons=1
> eq mu1: cons
>
> gllamm choice ed9-ed18 xvar2-xvar7 xvar56 acedu acexp acojt acoft, ///
>  i(id) base(5) link(mlogit) family(binom) constr(1 2 3 4) nip(2) ip(f) nrf(1) ///
>  eq(mu1) dots trace
>
>
>
>
> *
> *   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/
>



-- 
Stas Kolenikov, also found at http://stas.kolenikov.name
Small print: I use this email account for mailing lists only.

*
*   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/