st: Re: How to tackle the r1400 error message

[Nicola Man <nman1011@gmail.com>]

Hi,

Thanks for the suggestions.  The variables were mean centred (and the
variance couldn't be huge since the range of values were only from 1
to 7 at most).  I suspect the data structure might not be quite so
right for this, so I tried multinomial models with gllamm instead.  It
is now working ok with multinomial modelling, though I now have a
problem with getting intraclass correlations from them (will go in my
next message).

Regards,
Nicola

On Thu, Aug 26, 2010 at 7:38 PM, Nicola Man <nman1011@gmail.com> wrote:
> Hi,
>
> I am running into problems with numerical overflow according to the
> r(1400) error message, but really wanted to fit the multi-level model
> I am fitting.  It was working fine with the model fitting only random
> effects and intercept.
>
> When I added a few other fixed effects, it still managed to chuck out
> something (though not very sensible) when I ran xtmelogit with
> intpoints(1).  I then tried subsequently to fit with higher number of
> integration points.  This is when I start getting "initial values not
> feasible" with r(1400).
>
> I have STATA IC, and wonder if having SE or MP would help (though my
> impression from the list archive search tells me possibly not).
>
> If this does not work in STATA, then would anyone on this list know if
> SAS would?
>
> Help and advice appreciated!
>
> Regards,
> Nicola
>



-- 
Cheers,
Nicola

If urgent, please also email to:
n.man@unsw.edu.au
v2B2@hotmail.com

*
*   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/