Re: st: disattenuated regression

["'Alim Beveridge" <alimba@gmail.com>]

Hi,

Yes, the dichotomous variable (eg gender) is exogenous, and I did mean 
independent variable by IV not indicator (of a LV). I also want to 
include dummy variables for group membership as exogenous 
IVs/predictors. So if I understand you correctly, I don't need to worry 
about including these in my model and using standard ML estimation?
thanks,
'Alim


On 11/18/2012 11:51, JVerkuilen (Gmail) wrote:
> On Sun, Nov 18, 2012 at 11:47 AM, JVerkuilen (Gmail)
> <jvverkuilen@gmail.com>  wrote:
>    
> > On Sun, Nov 18, 2012 at 11:33 AM, 'Alim Beveridge<alimba@gmail.com>  wrote:
> >      
> > > Hi John,
> > > thanks again for your advice. I was thinking of doing this actually and have
> > > been teaching myself how to use sem in stata.
> > > However does it deal with dichotomous/categorical IVs, like Mplus does?
> > >        
> > If I understand your question correctly, your dichotomous variable is
> > exogenous, so you are fine. You're conditioning on it and thus its
> > distribution is not part of the likelihood to be optimized.
> >      
> Ah I think by IV you mean indicator variable, not independent variable.
>
> In that case, as long as you have an ordinal variable with a
> reasonable number of categories (more than 5) that has a reasonably
> symmetric bell shape, things are probably OK with ML, though you might
> want to use bootstrapping.
>
> Jay
> *
> *   For searches and help try:
> *   http://www.stata.com/help.cgi?search
> *   http://www.stata.com/support/faqs/resources/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/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/