I am not really sure it is relevant, but I think you should try
-gllamm- with what the authors call non-parametric likelihood
formulation.
On 1/23/06, Cecilia Frale <[email protected]> wrote:
> Dear Statalisters,
> here my problem, any suggestions are welcome.
>
> Let jit, (jit = 1, 2, 3), denote the survey response of firm i at time t.
> Suppose you are using a latent regression: yjit = 1 if �(j-1)i < yit* < �ji and
> 0 otherwise (j = 1, 2, 3), to deal with the categorical nature of data (we
> observe yjit instead of y*it ).
>
> Now consider that responses are related to a variable xt according to the
> conditional linear model yit = �i + �i xit where �i and �i are firm-specific
> time-invariant coefficients.
>
> I used a firm specific ordered probit model to estimate P(j|xt, i) , j = (1, 2,
> 3).
>
> Let f(xt) denote the time-invariant density function of xt. Therefore, the
> conditional probability of observing response j for firm i is P(j|i) = I(-inf,
> +inf)P(j|xt, i)f(xt)dxt , where I(-inf, +inf) is the integral over R.
>
> How can I calculate this integral using the INTEG Stata command?
>
>
>
> Cecilia
>
>
>
>
> *
> * 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/
>
--
Stas Kolenikov
http://stas.kolenikov.name
*
* 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/
