Possible problems
(1) The usual, update your Stata. -update query-
(2) Your initial parameters are infeasible, in particular 0.1
You can change that number or trying fitting with -areg-
(3) You linear model drops some variables, then
you don't have the same dimension in b0
(4) Your model is hard to fit then you shoudl try
-xttobit- but change the quadrature points, maybe you
model does not fit the theoretical assumptions.
R
----- Original Message -----
From: "Leonor Saravia" <[email protected]>
To: <[email protected]>
Sent: Tuesday, May 29, 2007 4:26 PM
Subject: st: xttobit with no feasible values
> Hello,
>
> I was trying to run a xttobit programm in Stata (using Stata 9) but it
> doesn't worked, in fact, it runs many hours and didn't converge to
> anything. So, now I'm trying to run the same programm (xttobit) but
> defining the initial values for the coefficients. To do this, I used
> the guide that David Drucker did
> (http://www.stata.com/statalist/archive/2003-05/msg00650.html). My
> problem is that when I run the programm Stata tells me that the
> initial values are not feasible when runs the xttobit. Please, if
> someone knows what's happening or how to solve the problem in other
> way I'd appreciated very much.
>
> What I did:
>
> First: I've estimated my model by OLS
>
> reg inversion98 tau region1-region12 EMBI industria1-industria8 DumIE
> desemp_r tam_emp1 tam_emp2
>
> Second: Generate the matrix of initial values
>
> matrix b0=e(b), ln(e(rmse)^2),.1
> matrix list b0
>
> Third: Run the random effects Tobit, using the "new" initial coefficients.
>
> xttobit inversion98 tau region1-region12 EMBI industria1-industria8
> DumIE desemp_r tam_emp1 tam_emp2, ll(0) from(b0, copy)
>
>
> Well, thank you very much for your time.
>
> Leonor
> *
> * 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/
*
* 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/
