Following what Marrten said - I concur. I have had this happen because
certain variables in my likelihood were evaluated as missing - usually
because of an coding or data organization error that led to trying to
evaluate a function outside of its domain. One way to find out what's
going on is to put a counter that increments each time the likelihood
program is called. Then you can stop after a few iterations (say 1-5),
then look at the contents of variables that are suppposed to contribute
to the evlaution of your likelihood. If necessary, you can create new
variables to show these values.
Al Feiveson
-----Original Message-----
From: [email protected]
[mailto:[email protected]] On Behalf Of Maarten buis
Sent: Thursday, October 30, 2008 6:52 AM
To: [email protected]
Subject: Re: st: Could not find feasible values(was: Syntax problems
with maximum likelihood (method lf))
This is typically an indication that there is an error in your
likelihood function. I know that this can be very frustrating but all we
can say is check that likelihood function again in detail. It is much
harder for us to do so as we don't know what the likelihood should be.
-- Maarten
--- Katia Bobulova <[email protected]> wrote:
> Dear all,
>
> thank you very much for your answers.
>
> I finally sorted out the problem with my program.
>
> However, when I typed ml check I received this message:
>
> The initial values are not feasible. This may be because the initial
> values have been chosen poorly or because there is an error in my
> program and it always returns missing no matter what the parameter
> values.
>
> Stata is going to use ml search to find a feasible set of initial
> values. If myprogram is broken, this will not work and you will have
> to press Break to make ml search stop.
> Searching...
> initial: log likelihood = -<inf> (could not be evaluated)
> searching for feasible values
> ..................................................
> could not find feasible values
> r(491);
>
> What should I do to modify myprogram so that it works? Is it possible
> that the problem is the number of observations, i.e. they are not
> enough?
>
> Could you please help me?I have really no idea of how to adjust my
> data.
>
> Thanks
> Katia
> *
> * 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/
>
-----------------------------------------
Maarten L. Buis
Department of Social Research Methodology Vrije Universiteit Amsterdam
Boelelaan 1081
1081 HV Amsterdam
The Netherlands
visiting address:
Buitenveldertselaan 3 (Metropolitan), room N515
+31 20 5986715
http://home.fsw.vu.nl/m.buis/
-----------------------------------------
*
* 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/
*
* 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/
